Determining usability differences between paid and free mobile health applications for cholesterol tracking

Background.;Heart disease due to high levels of cholesterol is the leading cause of death in the United States. It is therefore critical that people with heart disease or at risk for heart disease have tools that help to effectively manage cholesterol. Smartphones are portable information communication technologies that have the potential to help people manage their condition by making available software applications ("apps") that can help patients track their cholesterol throughout their activities of daily living. However, the degrees to which these cholesterol-tracking apps are usable may impact their effectiveness. A further consideration is that cholesterol tracking app usability may significantly differ according to pricing structure. I therefore set out to determine if paid and free smartphone apps differ in usability.;Methods.;The purpose of this study was to determine usability differences between paid and free smartphone apps that are designed to track cholesterol. I took multiple steps to generate a sample of cholesterol tracking apps: first I identified 153 apps in the Apple Computer, Inc., App StoreTM by searching on the keyword, "cholesterol"; I then selected a subset of 34 apps (19 paid and 15 free) based on data in seven pre-defined criteria: 1) patient-orientedness, 2) cholesterol level tracking capability, 3) star rating, 4) price, 5) app creation date, 6) app update date, and 7) and English language; from which I randomly selected 10 paid and 10 free apps. I determined usability scores for each for each app according to the Nielsen-Schneiderman Heuristic Framework, which estimates usability based on 14 criteria scored from 0 to 4 (a 0 indicates maximum usability and a 4 indicates minimal usability). To determine any usability differences, I used both descriptive and statistical comparisons including a two-tailed t-test and a Wilcoxon Rank-Sum test to determine differences with 95% confidence (alpha of .05).; Results.;My study found that when comparing the total average usability score of the paid apps and free apps, the paid app group was found to be more usable (1.44 vs. 1.64). However, not all paid apps were more usable than free apps. The free apps had better average usability scores in 4 out of the 14 usability categories. Median usability scores demonstrated an even smaller difference between paid and free apps. The paid apps had better median usability scores in 4 categories, the free apps had a better median usability scores in 2 categories, and scores were the same in the remaining 8 categories.;A 2-tailed Wilcoxon Rank-Sum test statistic for non-parametric data showed that there is no significant difference between the overall usability of paid and free apps (test statistic = 1.025, p-value of 0.3053). Similarly a 2 tailed t-test for parametric data also showed that there is no significant difference between the overall usability of paid and free apps (t-statistic --- 1.33, p-value of 0.18).;Both tests indicated the "language" variable significantly differed between paid and free apps: Wilcoxon Rank-Sum test (test statistic = 2.471, p-value = 0.0135), 2 tailed t-test (t-test statistic = 2.63, p-value = 0.017) thereby indicating that the paid.;Discussion.;This study demonstrated that except for the language variable, which indicates how easy the app is to understand, there was no significant difference in the usability of paid versus free apps. The results of this study suggest that paid apps can be better understood than free apps. When compared to the paid apps only one free app scored a zero rating (highest usability) for language.;mHealth app developers can successfully use this framework of to inform the usability of their apps. The strengths of this framework are that it is easy to follow and does require extensive resources to perform. This framework gives you defined criteria of usability for which to evaluate apps.;Conclusion.;There was no statistically significant difference between the overall usability of paid app instruction and free app instructions. The only usability criterion that was statistically significant area was language, meaning that, paid apps are easier to understand than free apps.