Nashat S. Moawad, MD

Unquestionably, using electronic information sources at the bedside has the potential to improve the quality of clinical care. That potential is moving closer to realization with FIRSTConsult (formerly PDxMD)—a multifunction physician resource designed for point-of-care use. Its name reflects its connection with MDConsult, the well-known electronic collection of textbooks. Both are produced by Elsevier, whose imprints include Saunders, Mosby, and Churchill Livingstone, and whose publications include Dorland’s Medical Dictionary, Gray’s Anatomy, Harriet Lane Handbook, and The Lancet.

FIRSTConsult comes in 2 incarnations: a Web-based version and a handheld version. The latter is available both for the Palm and Pocket PC operating systems, with variants of about 2, 6, and 10 MB. We evaluated the Web version (hereafter “FC-Web”) and the 10-MB Palm OS version (“FC-Palm”). Descriptions of the smaller versions are available on the FIRSTConsult web site.

As long as a user has access to a high-speed Internet connection, we give FC-Web a “thumbs up” as a rapid, real-time, point-of-care resource. FC-Palm is a portable mini-version of the Medical Conditions and Differential Diagnosis sections of FC-Web. In this article, we detail the advantages and disadvantages of both products.

FC-Web

FC-Web’s opening screen (Figure 1) lists its 5 major components, along with a search box that scans all components simultaneously. Searching “tinea,” for example, produces 529 retrievals (Figure 2). Users may modify search parameters with check boxes and drop-down lists.

FIGURE 1
FIRSTConsult’s opening screen

FIGURE 2
Typical FIRSTConsult search screen

Information retrieval with FC-Web

Clicking one of FC-Web’s components opens a search function or an alphabetic list to locate the desired monograph (Figure 3).

Improved search function. The search function for FC-Web is profoundly improved from that of its predecessor, PDxMD. The search retrievals include section titles (eg, Summary, Background, Diagnosis; see Figure 2), allowing users to link directly to the subject of interest. There is extensive mapping by synonyms, such that “AAA” finds “abdominal aortic aneurysm” and “TIA” finds “transient ischemic attack.” The search feature does not take wild cards or partial words and is intolerant to misspelling (eg, “vonWillebrand” vs “von Willebrand”).

Useful hyperlinks. FC-Web has extensive hyperlinks within its monographs and sections and to other components of the program, and they are among the most useful and logical links these reviewers have seen. Additionally, the program contains hyperlinks to MDConsult (for MDConsult subscribers only) and to excellent external resources, such as consensus statements. Where applicable, levels of evidence and references are cited.

Drawbacks. Regarding speed, we believe FC-Web requires high-speed access for true point-of-care use. FC-Web limits users to 1 open window—even hyperlinks to external sources cannot be opened in new windows.

FIGURE 3
Medical Conditions

Differential diagnosis section

This section provides differential diagnoses—broken into categories by patient age—for about 350 symptoms and signs. The search feature maps to close entries. For example, clicking on Differential Diagnosis and entering the search term “burping,” FC-Web maps to flatulence, eructation, and gas pain. Interestingly, some of the more common causes we see for burping in practice—anxiety/aerophagia, ingestion of dissolved gas (carbonated beverages), and ingestion of bicarbonate—do not appear.

For any condition found on the Differential Diagnosis monograph summary screens, a click pulls up a mini-summary of the illness, nicely leveraging the electronic environment (Figure 4). Users can click through to corresponding FC-Web disease monographs (if available) or to the topic in MDConsult (separate subscription required).

Limitations. Unlike differential diagnosis engines such as DXplain (available at www.lcs.mgh.harvard.edu/dxplain.htm), the Differential Diagnosis section of FC-Web does not allow users to enter sets of data (eg, age; sex; findings) to produce a list of possible diagnoses. Surprisingly, FC-web’s Differential section does not include “hyponatremia” (or “SIADH”), although it contains such things as “hypernasality” and “hypogeusia.”

FIGURE 4
Summary of a Differential Diagnosis

Patient handouts

FC-Web contains more than 450 patient education monographs.

The monographs have 3 main tabs. The first 2—View English and View Spanish—have drop-down menus offering handout choices of “Diagnosed Patient” and “General Information.” The third tab, Actions, allows users to customize handouts (add physician’s name, patient’s name, personalized instructions, etc). Defaults can be saved. Printouts are about 3 “airy” pages with liberal use of white space, large type, and bullet points. They are very basic and easy to read.

Handout deficiencies. Inexplicably, hepatitis A and B are included, but not hepatitis C, which would be the most useful information for our practices.

Occasional minor discrepancies exist between handouts and disease monographs. For example, the Acute Upper Respiratory Tract Infection handout says nothing about echinacea or zinc. However, the related monograph states “Echinacea and zinc supplements may be effective in reducing the duration and severity of common cold symptoms.”

Procedure files

Procedure Files contains 30 to 40 procedure monographs, including video clip illustrations of key portions; users have a choice of viewing these with Windows Media Player, RealPlayer, or QuickTime. Reference Centers contains 3 monographs: “Bioterrorism,” “Contraception,” and “Pregnancy” (as of May 2004). Although few in number, these Reference Centers are valuable point-of-care tools.

Medical conditions

We would expect Medical Conditions to be the workhorse section for most physicians. FC-Web advertises that it contains more than 450 conditions.

Rapid access to information. The best feature of this core section is its organization of information—eminently logical, extremely easy to navigate, bulleted, and extensively hyperlinked, taking full advantage of the e-medium.

To illustrate the outstanding layout and hyperlinking, if a user floats the cursor over a tab (eg, Diagnosis or Treatment) in a disease monograph, a menu of subsections drops down (Figure 5). Floating the cursor over some subsections produces a menu of additional options within that subsection. Furthermore, once selected, the resulting information is often extensively hyperlinked to related or more detailed information.

Generic drug names only. Throughout, FIRSTConsult sticks to generic drug names; if the user is uncertain what cefdinir or tamsulosin is, FIRSTConsult will offer no help.

Generous detail. In terms of content, Medical Conditions distills information into bullet points. FC-Web has more depth than, for example, Griffith’s 5-Minute Consult. Also, as a Web-based tool, FC-Web categorizes information multi-dimensionally with the horizontal main tabs plus drop-down, floating subcategories, allowing expeditious jumps to desired information. This organization facilitates rapid access to information.

Superb summary for point-of-care use. The Summary of Therapeutic Options subcategory of the Treatment section is particular noteworthy—an outstanding feature for busy clinicians—and concrete evidence that FC-Web was truly designed with point-of-care in mind. In this subsection, options are listed by bullet points and very briefly described (1 or 2 lines), which allows easy visual scanning. Each option (eg, medication) is then hyperlinked to its more detailed description (eg, dosage, route, etc) in the section, Drugs and Other Therapies: Details.

Sometimes, FC-Web fails to provide “drug of choice” recommendations, listing interventions randomly, which hinders rapid, point-of-care decision-making.

FIGURE 5
Typical FIRSTConsult monograph

Putting FC-Web to the test

FC-Web answered all general questions for which a monograph existed:

• What is the preferred therapy for head lice? (“Permethrin: … Treatment of choice for uncomplicated pediculosis.”)
• At what aneurysm diameter should a patient with an abdominal aortic aneurysm be referred? (“In general, patients with aneurysms larger than 5 cm in diameter, symptomatic aneurysms, or rapidly enlarging aneurysms should be considered for repair. Treatment for aneurysms between 4 and 5 cm is controversial.”)
• What is the basic workup for kidney stones?
• What are the indications for and monitoring of hydroxyurea for adults with sickle cell anemia?
• What are the therapeutic options for mycobacterium avium complex in an HIV-positive patient?
• How should one follow pulmonary sarcoid when tapering steroids? (A link to an external guideline was especially informative.)

Surprising shortcomings. FC-Web’s Medical Conditions section includes some less common entities—restless legs syndrome, Osgood-Schlatter’s disease, for example—but no monographs are devoted to conditions such as SIADH, paroxysmal supraventricular tachycardia, or portal or pulmonary hypertension.

Despite the evidence and reference citations, there were times we took issue with FC-Web. For example, FC-Web indicates the dosage of amoxicillin for otitis media in children older than 3 months is “25–45 mg/kg/day orally in divided doses (maximum dose 80 mg/kg/day).” In response to the rapid dissemination of drugresistant Streptococcus pneumoniae, the Centers for Disease Control has recommended, in general, that “the initial dose of amoxicillin should be increased from 40 to 45 mg/kg/day to 80 to 90 mg/kg/day.” Also, there are occasional disconnects within monographs. Under “Tinea Infections” in both the Summary and Diagnosis sections is a “Key! Don’t Miss Points” that says: “Patients with HIV infection may have severe onychomycosis and Malassezia folliculitis (pustular hair follicle infection).” However, no mention is made of how to treat Malassezia folliculitis.

Head-to-head with UpToDate. Compared with UpToDate [see “UpToDate: A comprehensive clinical database,” J Fam Pract 2003; 52(9):706–710], FC-Web obviously has been designed from a point-of-care focus. With UpToDate it is often laborious to cull key clinical information at the point-of-care. FC-Web is much less encyclopedic, has a superb layout, and whiz-click access to such amenities as the “Summary of therapeutic options”—with hyperlinking to details if needed.

Although FC-Web has no “history” or “favorites” functions, it is seldom a problem relocating information because of the simplicity of its design (unlike the complex paths one often follows in UpToDate to locate information).

FC-Web’s patient education monographs are more accessible, shorter, and at the opposite literacy extreme from UpToDate’s.

FC-Palm

FIRSTConsult versions for handhelds are available for download with purchase. On the Palm, users must first download and install FC’s EReader, a 90K application. Both EReader and FC-Palm easily load to memory cards and function briskly on recommended versions of Palm OS devices. However, “crashes” are common (and, in some instances, fully reproducible), requiring soft resets of the handheld device.

Shortcomings of the FC-Palm

FC-Palm’s opening screen lists its 2 components, Differential Diagnosis and Medical Conditions. (The other sections are not included in the handheld versions).

One noteworthy peculiarity with EReader is that, if the user presses the Palm’s off button, it exits the application but does not turn off the Palm. The user must press the button a second time to turn off the Palm. Similarly, at the user’s set “time out” (eg, after 2 minutes of inactivity), first FC-Palm/EReader times out, and only then does the countdown begin for the Palm to time out and turn off. Once users are forewarned, this is unlikely to be a problem, but we are not familiar with any other Palm application that exhibits this behavior.

Another curiosity: when FC-Palm times out, it does not resume where it exited. Users must retrace their steps to continue.

Levels of evidence are not included in FC-Palm. Elsevier indicates FC-Palm is updated approximately weekly. Unfortunately, there is no auto-update feature. Updating the 10MB version is time-consuming, particularly hot-syncing to a Palm card (and, if using dial-up connection, downloading), making it unlikely users would update frequently.

FC-Palm has no search function, and, interestingly, Palm’s search feature appears nonfunctional with this program.

Many medical handheld applications, ePocrates for example, employ lists that scroll alphabetically as users enter successive letters, such that entering “c,” “o,” and “l” would automatically scroll to the first listing under “c,” then under “co,” then under “col,” and so on. FC-Palm does not have this feature. Instead, it must be navigated by selecting a letter of the alphabet from the top of the screen (eg, R for Restless Legs), and then scrolling to the entry (via slide bar or Palm buttons) (Figure 6).

There are no icons or shortcuts to frequently used features (eg, “T” to jump to Therapy). Once in an entry, FC-Palm users navigate with a drop-down list to the section of interest. Each section is separate and can be entered only with the drop-down list (Figure 7).

No user-defined preferences are available to select FC-Palm’s monographs’ opening section (eg, Therapeutics).

Conveniently, FC-Palm has forward and back browser-type navigation controls.

Generic drug names are used throughout.

Comparing handheld versions, FIRSTConsult weighs in at 10 MB as opposed to UpToDate’s 800 MB—which may reflect upon the ease and speed of finding information at the point-of-care.

FIGURE 6
Navigating the Palm version

FIGURE 7
Selecting a topic in FC-Palm

Differential diagnosis

In content, FC-Palm’s Differential Diagnosis section is similar to that of FC-Web, except index listings appear only once (eg, “increased tearing, abnormal”) as opposed to appearing under all permutations in FC-Web’s index (eg, under “abnormal,” “increased,” and “tearing”). FC-Palm also employs extensive hyperlinking. Selecting “Incontinence, urinary” produces its differential; tapping “benign prostatic hypertrophy” in the resultant list produces a disease summary, with a link to the corresponding monograph in the Medical Conditions section.

The Differential Diagnosis section does not contain “hyponatremia,” but the Medical Conditions section does. Under the subheading Drugs and Other Therapies, only 2 therapies are listed—demeclocycline and fludrocortisone. There is no mention of water restriction (SIADH) or volume replacement (hypovolemia) per se—just drugs. In contrast, FC-Web’s comparable monograph incorporates volume replacement and free water restriction but only includes the issue of “urgent” volume replacement in the Immediate Action section, not in the important Summary of Therapeutic Options section.

Similarly, FC-Web’s monograph nicely summarizes drug options for pelvic inflammatory disease, but FC-Palm simply provides a listing of individual drugs without any hints about the appropriate combinations—significantly compromising users’ ability to quickly select therapy at point of care.

FC-Palm falls short of the mark for gonorrhea in the same way. The therapeutics screen simply lists 10 drugs (all hyperlinked to further information) without any hints as to recommended or preferred choices. In fact, hyperlinking to further information on ceftriaxone, the Centers for Disease Control’s recommended therapy in pregnancy, one finds the oxymoronic “Use caution in pregnancy (category B),” which might lead a neophyte on a search for the Holy Grail (ie, the nonexistent “better” drug for use in pregnancy).

Final word on firstconsult

FC-Web is well laid out and, using a high-speed connection, functional for point-of-care use, offering sufficiently detailed and extensively hyperlinked disease monographs. For the most common diseases, key questions are generally answered efficiently and in a manner consistent with current evidence or thought. As a bonus, FC-Web provides basic patient education handouts in English and Spanish.

FC-Palm lacks state-of-the-art navigation and search capabilities. FC-Web’s outstanding Summary of Therapeutic Options sections have been gutted in FC-Palm to be substantially less helpful—and potentially misleading—lists of non-prioritized therapies (primarily drugs).

Before we can recommend it as a prime-time handheld application, FC-Palm needs to upgrade its Therapeutics section to be equivalent to FC-Web’s Summary of Therapeutic Options; it should also remedy the nonstandard behaviors, improve navigation, add search options and some user preferences, and incorporate an auto-update feature.

Unquestionably, using electronic information sources at the bedside has the potential to improve the quality of clinical care. That potential is moving closer to realization with FIRSTConsult (formerly PDxMD)—a multifunction physician resource designed for point-of-care use. Its name reflects its connection with MDConsult, the well-known electronic collection of textbooks. Both are produced by Elsevier, whose imprints include Saunders, Mosby, and Churchill Livingstone, and whose publications include Dorland’s Medical Dictionary, Gray’s Anatomy, Harriet Lane Handbook, and The Lancet.

FIRSTConsult comes in 2 incarnations: a Web-based version and a handheld version. The latter is available both for the Palm and Pocket PC operating systems, with variants of about 2, 6, and 10 MB. We evaluated the Web version (hereafter “FC-Web”) and the 10-MB Palm OS version (“FC-Palm”). Descriptions of the smaller versions are available on the FIRSTConsult web site.

As long as a user has access to a high-speed Internet connection, we give FC-Web a “thumbs up” as a rapid, real-time, point-of-care resource. FC-Palm is a portable mini-version of the Medical Conditions and Differential Diagnosis sections of FC-Web. In this article, we detail the advantages and disadvantages of both products.

FC-Web

FC-Web’s opening screen (Figure 1) lists its 5 major components, along with a search box that scans all components simultaneously. Searching “tinea,” for example, produces 529 retrievals (Figure 2). Users may modify search parameters with check boxes and drop-down lists.

FIGURE 1
FIRSTConsult’s opening screen

FIGURE 2
Typical FIRSTConsult search screen

Information retrieval with FC-Web

Clicking one of FC-Web’s components opens a search function or an alphabetic list to locate the desired monograph (Figure 3).

Improved search function. The search function for FC-Web is profoundly improved from that of its predecessor, PDxMD. The search retrievals include section titles (eg, Summary, Background, Diagnosis; see Figure 2), allowing users to link directly to the subject of interest. There is extensive mapping by synonyms, such that “AAA” finds “abdominal aortic aneurysm” and “TIA” finds “transient ischemic attack.” The search feature does not take wild cards or partial words and is intolerant to misspelling (eg, “vonWillebrand” vs “von Willebrand”).

Useful hyperlinks. FC-Web has extensive hyperlinks within its monographs and sections and to other components of the program, and they are among the most useful and logical links these reviewers have seen. Additionally, the program contains hyperlinks to MDConsult (for MDConsult subscribers only) and to excellent external resources, such as consensus statements. Where applicable, levels of evidence and references are cited.

Drawbacks. Regarding speed, we believe FC-Web requires high-speed access for true point-of-care use. FC-Web limits users to 1 open window—even hyperlinks to external sources cannot be opened in new windows.

FIGURE 3
Medical Conditions

Differential diagnosis section

This section provides differential diagnoses—broken into categories by patient age—for about 350 symptoms and signs. The search feature maps to close entries. For example, clicking on Differential Diagnosis and entering the search term “burping,” FC-Web maps to flatulence, eructation, and gas pain. Interestingly, some of the more common causes we see for burping in practice—anxiety/aerophagia, ingestion of dissolved gas (carbonated beverages), and ingestion of bicarbonate—do not appear.

For any condition found on the Differential Diagnosis monograph summary screens, a click pulls up a mini-summary of the illness, nicely leveraging the electronic environment (Figure 4). Users can click through to corresponding FC-Web disease monographs (if available) or to the topic in MDConsult (separate subscription required).

Limitations. Unlike differential diagnosis engines such as DXplain (available at www.lcs.mgh.harvard.edu/dxplain.htm), the Differential Diagnosis section of FC-Web does not allow users to enter sets of data (eg, age; sex; findings) to produce a list of possible diagnoses. Surprisingly, FC-web’s Differential section does not include “hyponatremia” (or “SIADH”), although it contains such things as “hypernasality” and “hypogeusia.”

FIGURE 4
Summary of a Differential Diagnosis

Patient handouts

FC-Web contains more than 450 patient education monographs.

The monographs have 3 main tabs. The first 2—View English and View Spanish—have drop-down menus offering handout choices of “Diagnosed Patient” and “General Information.” The third tab, Actions, allows users to customize handouts (add physician’s name, patient’s name, personalized instructions, etc). Defaults can be saved. Printouts are about 3 “airy” pages with liberal use of white space, large type, and bullet points. They are very basic and easy to read.

Handout deficiencies. Inexplicably, hepatitis A and B are included, but not hepatitis C, which would be the most useful information for our practices.

Occasional minor discrepancies exist between handouts and disease monographs. For example, the Acute Upper Respiratory Tract Infection handout says nothing about echinacea or zinc. However, the related monograph states “Echinacea and zinc supplements may be effective in reducing the duration and severity of common cold symptoms.”

Procedure files

Procedure Files contains 30 to 40 procedure monographs, including video clip illustrations of key portions; users have a choice of viewing these with Windows Media Player, RealPlayer, or QuickTime. Reference Centers contains 3 monographs: “Bioterrorism,” “Contraception,” and “Pregnancy” (as of May 2004). Although few in number, these Reference Centers are valuable point-of-care tools.

Medical conditions

We would expect Medical Conditions to be the workhorse section for most physicians. FC-Web advertises that it contains more than 450 conditions.

Rapid access to information. The best feature of this core section is its organization of information—eminently logical, extremely easy to navigate, bulleted, and extensively hyperlinked, taking full advantage of the e-medium.

To illustrate the outstanding layout and hyperlinking, if a user floats the cursor over a tab (eg, Diagnosis or Treatment) in a disease monograph, a menu of subsections drops down (Figure 5). Floating the cursor over some subsections produces a menu of additional options within that subsection. Furthermore, once selected, the resulting information is often extensively hyperlinked to related or more detailed information.

Generic drug names only. Throughout, FIRSTConsult sticks to generic drug names; if the user is uncertain what cefdinir or tamsulosin is, FIRSTConsult will offer no help.

Generous detail. In terms of content, Medical Conditions distills information into bullet points. FC-Web has more depth than, for example, Griffith’s 5-Minute Consult. Also, as a Web-based tool, FC-Web categorizes information multi-dimensionally with the horizontal main tabs plus drop-down, floating subcategories, allowing expeditious jumps to desired information. This organization facilitates rapid access to information.

Superb summary for point-of-care use. The Summary of Therapeutic Options subcategory of the Treatment section is particular noteworthy—an outstanding feature for busy clinicians—and concrete evidence that FC-Web was truly designed with point-of-care in mind. In this subsection, options are listed by bullet points and very briefly described (1 or 2 lines), which allows easy visual scanning. Each option (eg, medication) is then hyperlinked to its more detailed description (eg, dosage, route, etc) in the section, Drugs and Other Therapies: Details.

Sometimes, FC-Web fails to provide “drug of choice” recommendations, listing interventions randomly, which hinders rapid, point-of-care decision-making.

FIGURE 5
Typical FIRSTConsult monograph

Putting FC-Web to the test

FC-Web answered all general questions for which a monograph existed:

• What is the preferred therapy for head lice? (“Permethrin: … Treatment of choice for uncomplicated pediculosis.”)
• At what aneurysm diameter should a patient with an abdominal aortic aneurysm be referred? (“In general, patients with aneurysms larger than 5 cm in diameter, symptomatic aneurysms, or rapidly enlarging aneurysms should be considered for repair. Treatment for aneurysms between 4 and 5 cm is controversial.”)
• What is the basic workup for kidney stones?
• What are the indications for and monitoring of hydroxyurea for adults with sickle cell anemia?
• What are the therapeutic options for mycobacterium avium complex in an HIV-positive patient?
• How should one follow pulmonary sarcoid when tapering steroids? (A link to an external guideline was especially informative.)

Surprising shortcomings. FC-Web’s Medical Conditions section includes some less common entities—restless legs syndrome, Osgood-Schlatter’s disease, for example—but no monographs are devoted to conditions such as SIADH, paroxysmal supraventricular tachycardia, or portal or pulmonary hypertension.

Despite the evidence and reference citations, there were times we took issue with FC-Web. For example, FC-Web indicates the dosage of amoxicillin for otitis media in children older than 3 months is “25–45 mg/kg/day orally in divided doses (maximum dose 80 mg/kg/day).” In response to the rapid dissemination of drugresistant Streptococcus pneumoniae, the Centers for Disease Control has recommended, in general, that “the initial dose of amoxicillin should be increased from 40 to 45 mg/kg/day to 80 to 90 mg/kg/day.” Also, there are occasional disconnects within monographs. Under “Tinea Infections” in both the Summary and Diagnosis sections is a “Key! Don’t Miss Points” that says: “Patients with HIV infection may have severe onychomycosis and Malassezia folliculitis (pustular hair follicle infection).” However, no mention is made of how to treat Malassezia folliculitis.

Head-to-head with UpToDate. Compared with UpToDate [see “UpToDate: A comprehensive clinical database,” J Fam Pract 2003; 52(9):706–710], FC-Web obviously has been designed from a point-of-care focus. With UpToDate it is often laborious to cull key clinical information at the point-of-care. FC-Web is much less encyclopedic, has a superb layout, and whiz-click access to such amenities as the “Summary of therapeutic options”—with hyperlinking to details if needed.

Although FC-Web has no “history” or “favorites” functions, it is seldom a problem relocating information because of the simplicity of its design (unlike the complex paths one often follows in UpToDate to locate information).

FC-Web’s patient education monographs are more accessible, shorter, and at the opposite literacy extreme from UpToDate’s.

FC-Palm

FIRSTConsult versions for handhelds are available for download with purchase. On the Palm, users must first download and install FC’s EReader, a 90K application. Both EReader and FC-Palm easily load to memory cards and function briskly on recommended versions of Palm OS devices. However, “crashes” are common (and, in some instances, fully reproducible), requiring soft resets of the handheld device.

Shortcomings of the FC-Palm

FC-Palm’s opening screen lists its 2 components, Differential Diagnosis and Medical Conditions. (The other sections are not included in the handheld versions).

One noteworthy peculiarity with EReader is that, if the user presses the Palm’s off button, it exits the application but does not turn off the Palm. The user must press the button a second time to turn off the Palm. Similarly, at the user’s set “time out” (eg, after 2 minutes of inactivity), first FC-Palm/EReader times out, and only then does the countdown begin for the Palm to time out and turn off. Once users are forewarned, this is unlikely to be a problem, but we are not familiar with any other Palm application that exhibits this behavior.

Another curiosity: when FC-Palm times out, it does not resume where it exited. Users must retrace their steps to continue.

Levels of evidence are not included in FC-Palm. Elsevier indicates FC-Palm is updated approximately weekly. Unfortunately, there is no auto-update feature. Updating the 10MB version is time-consuming, particularly hot-syncing to a Palm card (and, if using dial-up connection, downloading), making it unlikely users would update frequently.

FC-Palm has no search function, and, interestingly, Palm’s search feature appears nonfunctional with this program.

Many medical handheld applications, ePocrates for example, employ lists that scroll alphabetically as users enter successive letters, such that entering “c,” “o,” and “l” would automatically scroll to the first listing under “c,” then under “co,” then under “col,” and so on. FC-Palm does not have this feature. Instead, it must be navigated by selecting a letter of the alphabet from the top of the screen (eg, R for Restless Legs), and then scrolling to the entry (via slide bar or Palm buttons) (Figure 6).

There are no icons or shortcuts to frequently used features (eg, “T” to jump to Therapy). Once in an entry, FC-Palm users navigate with a drop-down list to the section of interest. Each section is separate and can be entered only with the drop-down list (Figure 7).

No user-defined preferences are available to select FC-Palm’s monographs’ opening section (eg, Therapeutics).

Conveniently, FC-Palm has forward and back browser-type navigation controls.

Generic drug names are used throughout.

Comparing handheld versions, FIRSTConsult weighs in at 10 MB as opposed to UpToDate’s 800 MB—which may reflect upon the ease and speed of finding information at the point-of-care.

FIGURE 6
Navigating the Palm version

FIGURE 7
Selecting a topic in FC-Palm

Differential diagnosis

In content, FC-Palm’s Differential Diagnosis section is similar to that of FC-Web, except index listings appear only once (eg, “increased tearing, abnormal”) as opposed to appearing under all permutations in FC-Web’s index (eg, under “abnormal,” “increased,” and “tearing”). FC-Palm also employs extensive hyperlinking. Selecting “Incontinence, urinary” produces its differential; tapping “benign prostatic hypertrophy” in the resultant list produces a disease summary, with a link to the corresponding monograph in the Medical Conditions section.

The Differential Diagnosis section does not contain “hyponatremia,” but the Medical Conditions section does. Under the subheading Drugs and Other Therapies, only 2 therapies are listed—demeclocycline and fludrocortisone. There is no mention of water restriction (SIADH) or volume replacement (hypovolemia) per se—just drugs. In contrast, FC-Web’s comparable monograph incorporates volume replacement and free water restriction but only includes the issue of “urgent” volume replacement in the Immediate Action section, not in the important Summary of Therapeutic Options section.

Similarly, FC-Web’s monograph nicely summarizes drug options for pelvic inflammatory disease, but FC-Palm simply provides a listing of individual drugs without any hints about the appropriate combinations—significantly compromising users’ ability to quickly select therapy at point of care.

FC-Palm falls short of the mark for gonorrhea in the same way. The therapeutics screen simply lists 10 drugs (all hyperlinked to further information) without any hints as to recommended or preferred choices. In fact, hyperlinking to further information on ceftriaxone, the Centers for Disease Control’s recommended therapy in pregnancy, one finds the oxymoronic “Use caution in pregnancy (category B),” which might lead a neophyte on a search for the Holy Grail (ie, the nonexistent “better” drug for use in pregnancy).

Final word on firstconsult

FC-Web is well laid out and, using a high-speed connection, functional for point-of-care use, offering sufficiently detailed and extensively hyperlinked disease monographs. For the most common diseases, key questions are generally answered efficiently and in a manner consistent with current evidence or thought. As a bonus, FC-Web provides basic patient education handouts in English and Spanish.

FC-Palm lacks state-of-the-art navigation and search capabilities. FC-Web’s outstanding Summary of Therapeutic Options sections have been gutted in FC-Palm to be substantially less helpful—and potentially misleading—lists of non-prioritized therapies (primarily drugs).

Before we can recommend it as a prime-time handheld application, FC-Palm needs to upgrade its Therapeutics section to be equivalent to FC-Web’s Summary of Therapeutic Options; it should also remedy the nonstandard behaviors, improve navigation, add search options and some user preferences, and incorporate an auto-update feature.

Unquestionably, using electronic information sources at the bedside has the potential to improve the quality of clinical care. That potential is moving closer to realization with FIRSTConsult (formerly PDxMD)—a multifunction physician resource designed for point-of-care use. Its name reflects its connection with MDConsult, the well-known electronic collection of textbooks. Both are produced by Elsevier, whose imprints include Saunders, Mosby, and Churchill Livingstone, and whose publications include Dorland’s Medical Dictionary, Gray’s Anatomy, Harriet Lane Handbook, and The Lancet.

FIRSTConsult comes in 2 incarnations: a Web-based version and a handheld version. The latter is available both for the Palm and Pocket PC operating systems, with variants of about 2, 6, and 10 MB. We evaluated the Web version (hereafter “FC-Web”) and the 10-MB Palm OS version (“FC-Palm”). Descriptions of the smaller versions are available on the FIRSTConsult web site.

As long as a user has access to a high-speed Internet connection, we give FC-Web a “thumbs up” as a rapid, real-time, point-of-care resource. FC-Palm is a portable mini-version of the Medical Conditions and Differential Diagnosis sections of FC-Web. In this article, we detail the advantages and disadvantages of both products.

FC-Web

FC-Web’s opening screen (Figure 1) lists its 5 major components, along with a search box that scans all components simultaneously. Searching “tinea,” for example, produces 529 retrievals (Figure 2). Users may modify search parameters with check boxes and drop-down lists.

FIGURE 1
FIRSTConsult’s opening screen

FIGURE 2
Typical FIRSTConsult search screen

Information retrieval with FC-Web

Clicking one of FC-Web’s components opens a search function or an alphabetic list to locate the desired monograph (Figure 3).

Improved search function. The search function for FC-Web is profoundly improved from that of its predecessor, PDxMD. The search retrievals include section titles (eg, Summary, Background, Diagnosis; see Figure 2), allowing users to link directly to the subject of interest. There is extensive mapping by synonyms, such that “AAA” finds “abdominal aortic aneurysm” and “TIA” finds “transient ischemic attack.” The search feature does not take wild cards or partial words and is intolerant to misspelling (eg, “vonWillebrand” vs “von Willebrand”).

Useful hyperlinks. FC-Web has extensive hyperlinks within its monographs and sections and to other components of the program, and they are among the most useful and logical links these reviewers have seen. Additionally, the program contains hyperlinks to MDConsult (for MDConsult subscribers only) and to excellent external resources, such as consensus statements. Where applicable, levels of evidence and references are cited.

Drawbacks. Regarding speed, we believe FC-Web requires high-speed access for true point-of-care use. FC-Web limits users to 1 open window—even hyperlinks to external sources cannot be opened in new windows.

FIGURE 3
Medical Conditions

Differential diagnosis section

This section provides differential diagnoses—broken into categories by patient age—for about 350 symptoms and signs. The search feature maps to close entries. For example, clicking on Differential Diagnosis and entering the search term “burping,” FC-Web maps to flatulence, eructation, and gas pain. Interestingly, some of the more common causes we see for burping in practice—anxiety/aerophagia, ingestion of dissolved gas (carbonated beverages), and ingestion of bicarbonate—do not appear.

For any condition found on the Differential Diagnosis monograph summary screens, a click pulls up a mini-summary of the illness, nicely leveraging the electronic environment (Figure 4). Users can click through to corresponding FC-Web disease monographs (if available) or to the topic in MDConsult (separate subscription required).

Limitations. Unlike differential diagnosis engines such as DXplain (available at www.lcs.mgh.harvard.edu/dxplain.htm), the Differential Diagnosis section of FC-Web does not allow users to enter sets of data (eg, age; sex; findings) to produce a list of possible diagnoses. Surprisingly, FC-web’s Differential section does not include “hyponatremia” (or “SIADH”), although it contains such things as “hypernasality” and “hypogeusia.”

FIGURE 4
Summary of a Differential Diagnosis

Patient handouts

FC-Web contains more than 450 patient education monographs.

The monographs have 3 main tabs. The first 2—View English and View Spanish—have drop-down menus offering handout choices of “Diagnosed Patient” and “General Information.” The third tab, Actions, allows users to customize handouts (add physician’s name, patient’s name, personalized instructions, etc). Defaults can be saved. Printouts are about 3 “airy” pages with liberal use of white space, large type, and bullet points. They are very basic and easy to read.

Handout deficiencies. Inexplicably, hepatitis A and B are included, but not hepatitis C, which would be the most useful information for our practices.

Occasional minor discrepancies exist between handouts and disease monographs. For example, the Acute Upper Respiratory Tract Infection handout says nothing about echinacea or zinc. However, the related monograph states “Echinacea and zinc supplements may be effective in reducing the duration and severity of common cold symptoms.”

Procedure files

Procedure Files contains 30 to 40 procedure monographs, including video clip illustrations of key portions; users have a choice of viewing these with Windows Media Player, RealPlayer, or QuickTime. Reference Centers contains 3 monographs: “Bioterrorism,” “Contraception,” and “Pregnancy” (as of May 2004). Although few in number, these Reference Centers are valuable point-of-care tools.

Medical conditions

We would expect Medical Conditions to be the workhorse section for most physicians. FC-Web advertises that it contains more than 450 conditions.

Rapid access to information. The best feature of this core section is its organization of information—eminently logical, extremely easy to navigate, bulleted, and extensively hyperlinked, taking full advantage of the e-medium.

To illustrate the outstanding layout and hyperlinking, if a user floats the cursor over a tab (eg, Diagnosis or Treatment) in a disease monograph, a menu of subsections drops down (Figure 5). Floating the cursor over some subsections produces a menu of additional options within that subsection. Furthermore, once selected, the resulting information is often extensively hyperlinked to related or more detailed information.

Generic drug names only. Throughout, FIRSTConsult sticks to generic drug names; if the user is uncertain what cefdinir or tamsulosin is, FIRSTConsult will offer no help.

Generous detail. In terms of content, Medical Conditions distills information into bullet points. FC-Web has more depth than, for example, Griffith’s 5-Minute Consult. Also, as a Web-based tool, FC-Web categorizes information multi-dimensionally with the horizontal main tabs plus drop-down, floating subcategories, allowing expeditious jumps to desired information. This organization facilitates rapid access to information.

Superb summary for point-of-care use. The Summary of Therapeutic Options subcategory of the Treatment section is particular noteworthy—an outstanding feature for busy clinicians—and concrete evidence that FC-Web was truly designed with point-of-care in mind. In this subsection, options are listed by bullet points and very briefly described (1 or 2 lines), which allows easy visual scanning. Each option (eg, medication) is then hyperlinked to its more detailed description (eg, dosage, route, etc) in the section, Drugs and Other Therapies: Details.

Sometimes, FC-Web fails to provide “drug of choice” recommendations, listing interventions randomly, which hinders rapid, point-of-care decision-making.

FIGURE 5
Typical FIRSTConsult monograph

Putting FC-Web to the test

FC-Web answered all general questions for which a monograph existed:

• What is the preferred therapy for head lice? (“Permethrin: … Treatment of choice for uncomplicated pediculosis.”)
• At what aneurysm diameter should a patient with an abdominal aortic aneurysm be referred? (“In general, patients with aneurysms larger than 5 cm in diameter, symptomatic aneurysms, or rapidly enlarging aneurysms should be considered for repair. Treatment for aneurysms between 4 and 5 cm is controversial.”)
• What is the basic workup for kidney stones?
• What are the indications for and monitoring of hydroxyurea for adults with sickle cell anemia?
• What are the therapeutic options for mycobacterium avium complex in an HIV-positive patient?
• How should one follow pulmonary sarcoid when tapering steroids? (A link to an external guideline was especially informative.)

Surprising shortcomings. FC-Web’s Medical Conditions section includes some less common entities—restless legs syndrome, Osgood-Schlatter’s disease, for example—but no monographs are devoted to conditions such as SIADH, paroxysmal supraventricular tachycardia, or portal or pulmonary hypertension.

Despite the evidence and reference citations, there were times we took issue with FC-Web. For example, FC-Web indicates the dosage of amoxicillin for otitis media in children older than 3 months is “25–45 mg/kg/day orally in divided doses (maximum dose 80 mg/kg/day).” In response to the rapid dissemination of drugresistant Streptococcus pneumoniae, the Centers for Disease Control has recommended, in general, that “the initial dose of amoxicillin should be increased from 40 to 45 mg/kg/day to 80 to 90 mg/kg/day.” Also, there are occasional disconnects within monographs. Under “Tinea Infections” in both the Summary and Diagnosis sections is a “Key! Don’t Miss Points” that says: “Patients with HIV infection may have severe onychomycosis and Malassezia folliculitis (pustular hair follicle infection).” However, no mention is made of how to treat Malassezia folliculitis.

Head-to-head with UpToDate. Compared with UpToDate [see “UpToDate: A comprehensive clinical database,” J Fam Pract 2003; 52(9):706–710], FC-Web obviously has been designed from a point-of-care focus. With UpToDate it is often laborious to cull key clinical information at the point-of-care. FC-Web is much less encyclopedic, has a superb layout, and whiz-click access to such amenities as the “Summary of therapeutic options”—with hyperlinking to details if needed.

Although FC-Web has no “history” or “favorites” functions, it is seldom a problem relocating information because of the simplicity of its design (unlike the complex paths one often follows in UpToDate to locate information).

FC-Web’s patient education monographs are more accessible, shorter, and at the opposite literacy extreme from UpToDate’s.

FC-Palm

FIRSTConsult versions for handhelds are available for download with purchase. On the Palm, users must first download and install FC’s EReader, a 90K application. Both EReader and FC-Palm easily load to memory cards and function briskly on recommended versions of Palm OS devices. However, “crashes” are common (and, in some instances, fully reproducible), requiring soft resets of the handheld device.

Shortcomings of the FC-Palm

FC-Palm’s opening screen lists its 2 components, Differential Diagnosis and Medical Conditions. (The other sections are not included in the handheld versions).

One noteworthy peculiarity with EReader is that, if the user presses the Palm’s off button, it exits the application but does not turn off the Palm. The user must press the button a second time to turn off the Palm. Similarly, at the user’s set “time out” (eg, after 2 minutes of inactivity), first FC-Palm/EReader times out, and only then does the countdown begin for the Palm to time out and turn off. Once users are forewarned, this is unlikely to be a problem, but we are not familiar with any other Palm application that exhibits this behavior.

Another curiosity: when FC-Palm times out, it does not resume where it exited. Users must retrace their steps to continue.

Levels of evidence are not included in FC-Palm. Elsevier indicates FC-Palm is updated approximately weekly. Unfortunately, there is no auto-update feature. Updating the 10MB version is time-consuming, particularly hot-syncing to a Palm card (and, if using dial-up connection, downloading), making it unlikely users would update frequently.

FC-Palm has no search function, and, interestingly, Palm’s search feature appears nonfunctional with this program.

Many medical handheld applications, ePocrates for example, employ lists that scroll alphabetically as users enter successive letters, such that entering “c,” “o,” and “l” would automatically scroll to the first listing under “c,” then under “co,” then under “col,” and so on. FC-Palm does not have this feature. Instead, it must be navigated by selecting a letter of the alphabet from the top of the screen (eg, R for Restless Legs), and then scrolling to the entry (via slide bar or Palm buttons) (Figure 6).

There are no icons or shortcuts to frequently used features (eg, “T” to jump to Therapy). Once in an entry, FC-Palm users navigate with a drop-down list to the section of interest. Each section is separate and can be entered only with the drop-down list (Figure 7).

No user-defined preferences are available to select FC-Palm’s monographs’ opening section (eg, Therapeutics).

Conveniently, FC-Palm has forward and back browser-type navigation controls.

Generic drug names are used throughout.

Comparing handheld versions, FIRSTConsult weighs in at 10 MB as opposed to UpToDate’s 800 MB—which may reflect upon the ease and speed of finding information at the point-of-care.

FIGURE 6
Navigating the Palm version

FIGURE 7
Selecting a topic in FC-Palm

Differential diagnosis

In content, FC-Palm’s Differential Diagnosis section is similar to that of FC-Web, except index listings appear only once (eg, “increased tearing, abnormal”) as opposed to appearing under all permutations in FC-Web’s index (eg, under “abnormal,” “increased,” and “tearing”). FC-Palm also employs extensive hyperlinking. Selecting “Incontinence, urinary” produces its differential; tapping “benign prostatic hypertrophy” in the resultant list produces a disease summary, with a link to the corresponding monograph in the Medical Conditions section.

The Differential Diagnosis section does not contain “hyponatremia,” but the Medical Conditions section does. Under the subheading Drugs and Other Therapies, only 2 therapies are listed—demeclocycline and fludrocortisone. There is no mention of water restriction (SIADH) or volume replacement (hypovolemia) per se—just drugs. In contrast, FC-Web’s comparable monograph incorporates volume replacement and free water restriction but only includes the issue of “urgent” volume replacement in the Immediate Action section, not in the important Summary of Therapeutic Options section.

Similarly, FC-Web’s monograph nicely summarizes drug options for pelvic inflammatory disease, but FC-Palm simply provides a listing of individual drugs without any hints about the appropriate combinations—significantly compromising users’ ability to quickly select therapy at point of care.

FC-Palm falls short of the mark for gonorrhea in the same way. The therapeutics screen simply lists 10 drugs (all hyperlinked to further information) without any hints as to recommended or preferred choices. In fact, hyperlinking to further information on ceftriaxone, the Centers for Disease Control’s recommended therapy in pregnancy, one finds the oxymoronic “Use caution in pregnancy (category B),” which might lead a neophyte on a search for the Holy Grail (ie, the nonexistent “better” drug for use in pregnancy).

Final word on firstconsult

FC-Web is well laid out and, using a high-speed connection, functional for point-of-care use, offering sufficiently detailed and extensively hyperlinked disease monographs. For the most common diseases, key questions are generally answered efficiently and in a manner consistent with current evidence or thought. As a bonus, FC-Web provides basic patient education handouts in English and Spanish.

FC-Palm lacks state-of-the-art navigation and search capabilities. FC-Web’s outstanding Summary of Therapeutic Options sections have been gutted in FC-Palm to be substantially less helpful—and potentially misleading—lists of non-prioritized therapies (primarily drugs).

Before we can recommend it as a prime-time handheld application, FC-Palm needs to upgrade its Therapeutics section to be equivalent to FC-Web’s Summary of Therapeutic Options; it should also remedy the nonstandard behaviors, improve navigation, add search options and some user preferences, and incorporate an auto-update feature.

How many clinical questions arise during an average day at the office? How many more questions might one entertain if clear-cut, expert advice were seconds away? Enter UpToDate, an electronic clinical database for, well, acquiring up-to-date clinical information—plus continuing medical education (CME) credits and patient handouts.

Unlike a textbook, UpToDate is not set up for paging through content from beginning to end. Rather, it is designed for searches that yield lists of relevant monographs. UpToDate’s initial screen displays simple menu bars and a prominent search box (Figure 1). The next step allows one to narrow the search (Figure 2).

Per its educational objectives, UpToDate provides “instant access to the most current clinical information and recommendations on patient care and treatment,” yet is “comprehensive” from “physiological concepts through results of the most recent clinical trials and protocols.” Each monograph indicates when the literature was last reviewed and the document updated. The service is available in 3 formats: personal computer-based, online, and Pocket PC. We reviewed these formats using Windows PCs.

FIGURE 1
UpToDate’s main screen

FIGURE 2
Typical UpToDate search screen

Testing uptodate in a clinical context

Following are a few questions that needed answering during a half-day of practice:

• How effective is topical terbinafine (Lamisil) in treating Candida skin infections?
• What is a reasonable algorithm to determine the cause of chronic renal failure?
• How does one compound nitroglycerin ointment to treat an anal fissure?
• What is the latest thinking on interstitial cystitis?
• What is the preferred empiric antibiotic therapy for presumed bacterial conjunctivitis in adults?
• Does efficacy differ between injectable and intranasal forms of calcitonin for relief of pain from osteoporotic vertebral compression fractures in postmenopausal women?

UpToDate answered 4 of the 6 questions quickly, requiring no more than a few minutes for each search. We found that 2% nitroglycerin must be diluted to 0.2% for use on anal fissures. We also obtained summarized information on interstitial cystitis, a treatment recommendation for conjunctivitis, and data regarding the efficacy of calcitonin administered by either of 2 routes. The information generally was excellent.

More difficult searches

Testing UpToDate further, we searched for less common maladies. We found 4 sentences on proctalgia fugax that, remarkably, seemed to cover the topic adequately. Transient global amnesia was covered tersely (3 sentences) but, again, adequately. Restless legs syndrome was covered well, in several paragraphs, although drug therapies were discussed in random order rather than by drug of choice. Piriformis syndrome was covered inadequately, and presbyesophagus was not listed at all.

Controversial topics

Subsequent searches yielded excellent discussions on controversial topics. Many monographs lend themselves best to leisurely reading, but they could also be used at the point of care. For example, the controversial diastolic “J-curve phenomenon” in treating hypertension is covered in about fourteen 2- to 6-sentence paragraphs. A summary appears first, typical of UpToDate’s style (Figure 3).

This section on diastolic blood pressure is subdivided into “Positive studies,” “Negative studies,” and the “HOT trial.” Users may jump directly to the “HOT trial” summary by clicking on that title on the left window. References appear in a text box at the bottom of the screen. The relative sizes of these 3 windows may be changed by grabbing and dragging the lines dividing them.

FIGURE 3
Typical UpToDate monograph

Drug mentions

Drugs mentioned in UpToDate are hot-linked to the corresponding Lexi-Comp drug information monograph, which is “intended to serve as a rapidly accessible, concise initial [drug] reference resource.” UpToDate/Lexi-Comp does not provide guidance regarding cost of drug therapies.

Notable Features

Graphics

When the user clicks on a graphic’s link in the text, a figure, table, or picture is superimposed on the text. The image disappears when the user returns to the text.

Patient education handouts

UpToDate offers patient education handouts; however, these could be made much more accessible. Veteran users know exactly where to look for these handouts, but novices may apply the “3 clicks and you’re out” rule when the process proves too frustrating.

Both “Prostate Cancer Screening” (including PSA testing) and “Screening for Breast Cancer” (including mammography) are about 5 single-spaced pages long when printed using Times New Roman 12-point. Although controversies are covered well, our guesstimate of the reading comprehension level is 1 or 2 years post–high school— maybe too difficult for patient education.

Search engine

UpToDate’s search engine is rudimentary. Multiword medical term searches are supported (eg, “chronic renal failure” and “vertebral compression fracture”), as are initial word fragments (eg, “spondyl”). However, searches incorporating Boolean operators (AND, OR, NOT) and unrelated words are not directly supported, although an obtuse “secondary search” does provide for one AND operation. On this secondary search screen, users may choose to focus on a specific category of information, such as diagnosis, treatment, pathogenesis, or patient information.

The search function can be quirky. Although the words “gluteus medius bursitis” appear in the text, cutting and pasting those words into the search function yields no retrievals; this makes it more difficult to locate information. Searching “restless legs syndrome” produces the suggestion that the user search “restless legs”—which, when searched, produces a most relevant topic: “restless legs syndrome.”

Additional digital tools

Other features of UpToDate take excellent advantage of the electronic environment.

Hyperlinking. Extensive hyperlinking within and between monographs facilitates access to information.

Retracing steps. “Back” and “forward” buttons allow users to navigate linearly.

Text-string searches.The “Find in Text” feature, or its keyboard shortcut, Ctrl-F, allows users to find text string matches within the current monograph.

Search history. A “Recent Topics” button in the menu bar makes it possible to revisit the prior 20 topics viewed by a user.

Accessing citations. Clicking on any reference links users to that article’s citation and, if available, MEDLINE abstract. Citations may be exported to a text file, but may not be directly cut-and-pasted.

Exporting monographs. Users may print a monograph, export a monograph as text or in rich text format (*.rtf), export a monograph’s abstracts into WordPad, or (from the Web version) e-mail a monograph to a colleague with an optional, complimentary 7-day guest pass to UpToDate Online. Graphics can be exported into Windows Paint.

CME tracking. UpToDate keeps track of a user’s “CME time,” crediting up to 10 minutes (in hundredths of an hour increments) per topic per use of UpToDate. CME from the CD is unlimited; up to 50 hours may be accumulated online. Credits may be totalled for submission for a fee of $25 per transmittal.

Room for improvement

Some of UpToDate’s advice is open to question. Although information is generally well-referenced, a few statements should have been referenced but were not.

For example (in reference to bacterial conjunctivitis): “Aminoglycoside drops and ointments (Genoptic^®, Tobrex^®) are poor choices since they are toxic to the corneal epithelium and can cause a reactive keratoconjunctivitis after several days of use.”

UpToDate prominently advises that the Pediatric module is a work in progress. In searching for “delayed menarche,” the closest match we found was “Pubertal changes (Tanner stages).” Information was sufficient to answer most clinical questions. However, UpToDate did not provide a clinically useful graphic listing sentinel events (Tanner stages, peak height velocity, menarche) in typical chronological order with anticipated normal age ranges.

Digital design problems

UpToDate fails to take full advantage of the electronic environment. Inexplicably, in the Contents menu screens, Find in Text is not active, and users cannot navigate using the Page-Up/Page-Down or (Ctrl+)Home/End buttons on their keypads. For example, following Contents through “Patient Education” to “Women’s Health Issues,” users find 48 topics in 19 sections as they try to locate the one of interest.

“Mammography” is integrated into “Screening for breast cancer,” and is not a separate keyword for users who are visually scanning for that word (the only option).

Right-clicking is not supported.

An Editmenu is absent. Windows’ Cut and Paste keyboard conventions (Ctrl-c/Ctrl-v) are supported in the main UpToDate window—but not in other windows.

Bookmarking—the electronic equivalent of the folded page corner—is unavailable, as is a favoriteslist.

Highlighting—electronic yellow, blue, and pink markers—is not supported.

No facility for user-added annotation is available.

Opening multiple windows of information simultaneously is not possible.

Standard resizing of the main UpToDate window (by “grabbing” edges of the window) is not supported.

Finally, an ability to limit searches from the main screen to patient handouts or Lexi-Comp would be desirable.

Pocket version

UpToDate also provides a Pocket PC version. The Pocket PC version searches the same database with a search engine similar to the other versions. It lacks the search history offered in the CD version. Although faster than searching through a textbook, an average UpToDate search on Dell Axim X5 400 mHz (Intel PXA250) with 64 MB RAM and a SanDisk 1G CompactFlash card took about 10 to 27 seconds, plus more time to navigate through the long and sometimes cumbersome submenus.

The primary intent for palmtop medical software is point-of-care use. The extensiveness of UpToDate’s text—occupying 800 MB of the 1G card—and limited speed of palmtop computers often makes this impractical. However, it should be emphasized that this rich, inclusive database is a valuable reference when time is not at a premium. Certainly, the Pocket PC version provides ultimate portability. Users contemplating use of UpToDate/Pocket PC should budget for a highend Pocket PC and a 1G card.

Conclusions

UpToDate is a clinically useful, searchable database of medical information updated every 4 months and available on CD, online, and for souped-up Pocket PC palmtop devices. It is almost as if UpToDate purposely incorporated an unsophisticated search engine and did not optimize the electronic environment to flaunt its forte—the clinical database.

For family physicians, the database is of sufficiently high quality, and it will be even more useful as the pediatric module expands. We hope that future enhancements to UpToDate will include clinical bullets for point-of-care use and optimization of the electronic environment. The latter should include implementation of Boolean searches; bookmarks; highlighters; a user-added note facility; copying and pasting of references; links to drug cost information; and—the whole point of Windows! — multi-window capability and conventional window resizing.

Patient information needs to be more accessible to be clinically useful. Also, versions of the patient information handouts with less complex language to appeal to the average patient would be helpful. As is, UpToDate’s importance will only increase. Highly recommended.

Correspondence
Gary N. Fox, MD, Mercy Health Partners Family Practice Residency Program, 2200 Jefferson Ave, Toledo OH 43624-1117. E-mail: foxgary@yahoo.com.

How many clinical questions arise during an average day at the office? How many more questions might one entertain if clear-cut, expert advice were seconds away? Enter UpToDate, an electronic clinical database for, well, acquiring up-to-date clinical information—plus continuing medical education (CME) credits and patient handouts.

Unlike a textbook, UpToDate is not set up for paging through content from beginning to end. Rather, it is designed for searches that yield lists of relevant monographs. UpToDate’s initial screen displays simple menu bars and a prominent search box (Figure 1). The next step allows one to narrow the search (Figure 2).

Per its educational objectives, UpToDate provides “instant access to the most current clinical information and recommendations on patient care and treatment,” yet is “comprehensive” from “physiological concepts through results of the most recent clinical trials and protocols.” Each monograph indicates when the literature was last reviewed and the document updated. The service is available in 3 formats: personal computer-based, online, and Pocket PC. We reviewed these formats using Windows PCs.

FIGURE 1
UpToDate’s main screen

FIGURE 2
Typical UpToDate search screen

Testing uptodate in a clinical context

Following are a few questions that needed answering during a half-day of practice:

• How effective is topical terbinafine (Lamisil) in treating Candida skin infections?
• What is a reasonable algorithm to determine the cause of chronic renal failure?
• How does one compound nitroglycerin ointment to treat an anal fissure?
• What is the latest thinking on interstitial cystitis?
• What is the preferred empiric antibiotic therapy for presumed bacterial conjunctivitis in adults?
• Does efficacy differ between injectable and intranasal forms of calcitonin for relief of pain from osteoporotic vertebral compression fractures in postmenopausal women?

UpToDate answered 4 of the 6 questions quickly, requiring no more than a few minutes for each search. We found that 2% nitroglycerin must be diluted to 0.2% for use on anal fissures. We also obtained summarized information on interstitial cystitis, a treatment recommendation for conjunctivitis, and data regarding the efficacy of calcitonin administered by either of 2 routes. The information generally was excellent.

More difficult searches

Testing UpToDate further, we searched for less common maladies. We found 4 sentences on proctalgia fugax that, remarkably, seemed to cover the topic adequately. Transient global amnesia was covered tersely (3 sentences) but, again, adequately. Restless legs syndrome was covered well, in several paragraphs, although drug therapies were discussed in random order rather than by drug of choice. Piriformis syndrome was covered inadequately, and presbyesophagus was not listed at all.

Controversial topics

Subsequent searches yielded excellent discussions on controversial topics. Many monographs lend themselves best to leisurely reading, but they could also be used at the point of care. For example, the controversial diastolic “J-curve phenomenon” in treating hypertension is covered in about fourteen 2- to 6-sentence paragraphs. A summary appears first, typical of UpToDate’s style (Figure 3).

This section on diastolic blood pressure is subdivided into “Positive studies,” “Negative studies,” and the “HOT trial.” Users may jump directly to the “HOT trial” summary by clicking on that title on the left window. References appear in a text box at the bottom of the screen. The relative sizes of these 3 windows may be changed by grabbing and dragging the lines dividing them.

FIGURE 3
Typical UpToDate monograph

Drug mentions

Drugs mentioned in UpToDate are hot-linked to the corresponding Lexi-Comp drug information monograph, which is “intended to serve as a rapidly accessible, concise initial [drug] reference resource.” UpToDate/Lexi-Comp does not provide guidance regarding cost of drug therapies.

Notable Features

Graphics

When the user clicks on a graphic’s link in the text, a figure, table, or picture is superimposed on the text. The image disappears when the user returns to the text.

Patient education handouts

UpToDate offers patient education handouts; however, these could be made much more accessible. Veteran users know exactly where to look for these handouts, but novices may apply the “3 clicks and you’re out” rule when the process proves too frustrating.

Both “Prostate Cancer Screening” (including PSA testing) and “Screening for Breast Cancer” (including mammography) are about 5 single-spaced pages long when printed using Times New Roman 12-point. Although controversies are covered well, our guesstimate of the reading comprehension level is 1 or 2 years post–high school— maybe too difficult for patient education.

Search engine

UpToDate’s search engine is rudimentary. Multiword medical term searches are supported (eg, “chronic renal failure” and “vertebral compression fracture”), as are initial word fragments (eg, “spondyl”). However, searches incorporating Boolean operators (AND, OR, NOT) and unrelated words are not directly supported, although an obtuse “secondary search” does provide for one AND operation. On this secondary search screen, users may choose to focus on a specific category of information, such as diagnosis, treatment, pathogenesis, or patient information.

The search function can be quirky. Although the words “gluteus medius bursitis” appear in the text, cutting and pasting those words into the search function yields no retrievals; this makes it more difficult to locate information. Searching “restless legs syndrome” produces the suggestion that the user search “restless legs”—which, when searched, produces a most relevant topic: “restless legs syndrome.”

Additional digital tools

Other features of UpToDate take excellent advantage of the electronic environment.

Hyperlinking. Extensive hyperlinking within and between monographs facilitates access to information.

Retracing steps. “Back” and “forward” buttons allow users to navigate linearly.

Text-string searches.The “Find in Text” feature, or its keyboard shortcut, Ctrl-F, allows users to find text string matches within the current monograph.

Search history. A “Recent Topics” button in the menu bar makes it possible to revisit the prior 20 topics viewed by a user.

Accessing citations. Clicking on any reference links users to that article’s citation and, if available, MEDLINE abstract. Citations may be exported to a text file, but may not be directly cut-and-pasted.

Exporting monographs. Users may print a monograph, export a monograph as text or in rich text format (*.rtf), export a monograph’s abstracts into WordPad, or (from the Web version) e-mail a monograph to a colleague with an optional, complimentary 7-day guest pass to UpToDate Online. Graphics can be exported into Windows Paint.

CME tracking. UpToDate keeps track of a user’s “CME time,” crediting up to 10 minutes (in hundredths of an hour increments) per topic per use of UpToDate. CME from the CD is unlimited; up to 50 hours may be accumulated online. Credits may be totalled for submission for a fee of $25 per transmittal.

Room for improvement

Some of UpToDate’s advice is open to question. Although information is generally well-referenced, a few statements should have been referenced but were not.

For example (in reference to bacterial conjunctivitis): “Aminoglycoside drops and ointments (Genoptic^®, Tobrex^®) are poor choices since they are toxic to the corneal epithelium and can cause a reactive keratoconjunctivitis after several days of use.”

UpToDate prominently advises that the Pediatric module is a work in progress. In searching for “delayed menarche,” the closest match we found was “Pubertal changes (Tanner stages).” Information was sufficient to answer most clinical questions. However, UpToDate did not provide a clinically useful graphic listing sentinel events (Tanner stages, peak height velocity, menarche) in typical chronological order with anticipated normal age ranges.

Digital design problems

UpToDate fails to take full advantage of the electronic environment. Inexplicably, in the Contents menu screens, Find in Text is not active, and users cannot navigate using the Page-Up/Page-Down or (Ctrl+)Home/End buttons on their keypads. For example, following Contents through “Patient Education” to “Women’s Health Issues,” users find 48 topics in 19 sections as they try to locate the one of interest.

“Mammography” is integrated into “Screening for breast cancer,” and is not a separate keyword for users who are visually scanning for that word (the only option).

Right-clicking is not supported.

An Editmenu is absent. Windows’ Cut and Paste keyboard conventions (Ctrl-c/Ctrl-v) are supported in the main UpToDate window—but not in other windows.

Bookmarking—the electronic equivalent of the folded page corner—is unavailable, as is a favoriteslist.

Highlighting—electronic yellow, blue, and pink markers—is not supported.

No facility for user-added annotation is available.

Opening multiple windows of information simultaneously is not possible.

Standard resizing of the main UpToDate window (by “grabbing” edges of the window) is not supported.

Finally, an ability to limit searches from the main screen to patient handouts or Lexi-Comp would be desirable.

Pocket version

UpToDate also provides a Pocket PC version. The Pocket PC version searches the same database with a search engine similar to the other versions. It lacks the search history offered in the CD version. Although faster than searching through a textbook, an average UpToDate search on Dell Axim X5 400 mHz (Intel PXA250) with 64 MB RAM and a SanDisk 1G CompactFlash card took about 10 to 27 seconds, plus more time to navigate through the long and sometimes cumbersome submenus.

The primary intent for palmtop medical software is point-of-care use. The extensiveness of UpToDate’s text—occupying 800 MB of the 1G card—and limited speed of palmtop computers often makes this impractical. However, it should be emphasized that this rich, inclusive database is a valuable reference when time is not at a premium. Certainly, the Pocket PC version provides ultimate portability. Users contemplating use of UpToDate/Pocket PC should budget for a highend Pocket PC and a 1G card.

Conclusions

UpToDate is a clinically useful, searchable database of medical information updated every 4 months and available on CD, online, and for souped-up Pocket PC palmtop devices. It is almost as if UpToDate purposely incorporated an unsophisticated search engine and did not optimize the electronic environment to flaunt its forte—the clinical database.

For family physicians, the database is of sufficiently high quality, and it will be even more useful as the pediatric module expands. We hope that future enhancements to UpToDate will include clinical bullets for point-of-care use and optimization of the electronic environment. The latter should include implementation of Boolean searches; bookmarks; highlighters; a user-added note facility; copying and pasting of references; links to drug cost information; and—the whole point of Windows! — multi-window capability and conventional window resizing.

Patient information needs to be more accessible to be clinically useful. Also, versions of the patient information handouts with less complex language to appeal to the average patient would be helpful. As is, UpToDate’s importance will only increase. Highly recommended.

Correspondence
Gary N. Fox, MD, Mercy Health Partners Family Practice Residency Program, 2200 Jefferson Ave, Toledo OH 43624-1117. E-mail: foxgary@yahoo.com.

How many clinical questions arise during an average day at the office? How many more questions might one entertain if clear-cut, expert advice were seconds away? Enter UpToDate, an electronic clinical database for, well, acquiring up-to-date clinical information—plus continuing medical education (CME) credits and patient handouts.

Unlike a textbook, UpToDate is not set up for paging through content from beginning to end. Rather, it is designed for searches that yield lists of relevant monographs. UpToDate’s initial screen displays simple menu bars and a prominent search box (Figure 1). The next step allows one to narrow the search (Figure 2).

Per its educational objectives, UpToDate provides “instant access to the most current clinical information and recommendations on patient care and treatment,” yet is “comprehensive” from “physiological concepts through results of the most recent clinical trials and protocols.” Each monograph indicates when the literature was last reviewed and the document updated. The service is available in 3 formats: personal computer-based, online, and Pocket PC. We reviewed these formats using Windows PCs.

FIGURE 1
UpToDate’s main screen

FIGURE 2
Typical UpToDate search screen

Testing uptodate in a clinical context

Following are a few questions that needed answering during a half-day of practice:

• How effective is topical terbinafine (Lamisil) in treating Candida skin infections?
• What is a reasonable algorithm to determine the cause of chronic renal failure?
• How does one compound nitroglycerin ointment to treat an anal fissure?
• What is the latest thinking on interstitial cystitis?
• What is the preferred empiric antibiotic therapy for presumed bacterial conjunctivitis in adults?
• Does efficacy differ between injectable and intranasal forms of calcitonin for relief of pain from osteoporotic vertebral compression fractures in postmenopausal women?

UpToDate answered 4 of the 6 questions quickly, requiring no more than a few minutes for each search. We found that 2% nitroglycerin must be diluted to 0.2% for use on anal fissures. We also obtained summarized information on interstitial cystitis, a treatment recommendation for conjunctivitis, and data regarding the efficacy of calcitonin administered by either of 2 routes. The information generally was excellent.

More difficult searches

Testing UpToDate further, we searched for less common maladies. We found 4 sentences on proctalgia fugax that, remarkably, seemed to cover the topic adequately. Transient global amnesia was covered tersely (3 sentences) but, again, adequately. Restless legs syndrome was covered well, in several paragraphs, although drug therapies were discussed in random order rather than by drug of choice. Piriformis syndrome was covered inadequately, and presbyesophagus was not listed at all.

Controversial topics

Subsequent searches yielded excellent discussions on controversial topics. Many monographs lend themselves best to leisurely reading, but they could also be used at the point of care. For example, the controversial diastolic “J-curve phenomenon” in treating hypertension is covered in about fourteen 2- to 6-sentence paragraphs. A summary appears first, typical of UpToDate’s style (Figure 3).

This section on diastolic blood pressure is subdivided into “Positive studies,” “Negative studies,” and the “HOT trial.” Users may jump directly to the “HOT trial” summary by clicking on that title on the left window. References appear in a text box at the bottom of the screen. The relative sizes of these 3 windows may be changed by grabbing and dragging the lines dividing them.

FIGURE 3
Typical UpToDate monograph

Drug mentions

Drugs mentioned in UpToDate are hot-linked to the corresponding Lexi-Comp drug information monograph, which is “intended to serve as a rapidly accessible, concise initial [drug] reference resource.” UpToDate/Lexi-Comp does not provide guidance regarding cost of drug therapies.

Notable Features

Graphics

When the user clicks on a graphic’s link in the text, a figure, table, or picture is superimposed on the text. The image disappears when the user returns to the text.

Patient education handouts

UpToDate offers patient education handouts; however, these could be made much more accessible. Veteran users know exactly where to look for these handouts, but novices may apply the “3 clicks and you’re out” rule when the process proves too frustrating.

Both “Prostate Cancer Screening” (including PSA testing) and “Screening for Breast Cancer” (including mammography) are about 5 single-spaced pages long when printed using Times New Roman 12-point. Although controversies are covered well, our guesstimate of the reading comprehension level is 1 or 2 years post–high school— maybe too difficult for patient education.

Search engine

UpToDate’s search engine is rudimentary. Multiword medical term searches are supported (eg, “chronic renal failure” and “vertebral compression fracture”), as are initial word fragments (eg, “spondyl”). However, searches incorporating Boolean operators (AND, OR, NOT) and unrelated words are not directly supported, although an obtuse “secondary search” does provide for one AND operation. On this secondary search screen, users may choose to focus on a specific category of information, such as diagnosis, treatment, pathogenesis, or patient information.

The search function can be quirky. Although the words “gluteus medius bursitis” appear in the text, cutting and pasting those words into the search function yields no retrievals; this makes it more difficult to locate information. Searching “restless legs syndrome” produces the suggestion that the user search “restless legs”—which, when searched, produces a most relevant topic: “restless legs syndrome.”

Additional digital tools

Other features of UpToDate take excellent advantage of the electronic environment.

Hyperlinking. Extensive hyperlinking within and between monographs facilitates access to information.

Retracing steps. “Back” and “forward” buttons allow users to navigate linearly.

Text-string searches.The “Find in Text” feature, or its keyboard shortcut, Ctrl-F, allows users to find text string matches within the current monograph.

Search history. A “Recent Topics” button in the menu bar makes it possible to revisit the prior 20 topics viewed by a user.

Accessing citations. Clicking on any reference links users to that article’s citation and, if available, MEDLINE abstract. Citations may be exported to a text file, but may not be directly cut-and-pasted.

Exporting monographs. Users may print a monograph, export a monograph as text or in rich text format (*.rtf), export a monograph’s abstracts into WordPad, or (from the Web version) e-mail a monograph to a colleague with an optional, complimentary 7-day guest pass to UpToDate Online. Graphics can be exported into Windows Paint.

CME tracking. UpToDate keeps track of a user’s “CME time,” crediting up to 10 minutes (in hundredths of an hour increments) per topic per use of UpToDate. CME from the CD is unlimited; up to 50 hours may be accumulated online. Credits may be totalled for submission for a fee of $25 per transmittal.

Room for improvement

Some of UpToDate’s advice is open to question. Although information is generally well-referenced, a few statements should have been referenced but were not.

For example (in reference to bacterial conjunctivitis): “Aminoglycoside drops and ointments (Genoptic^®, Tobrex^®) are poor choices since they are toxic to the corneal epithelium and can cause a reactive keratoconjunctivitis after several days of use.”

UpToDate prominently advises that the Pediatric module is a work in progress. In searching for “delayed menarche,” the closest match we found was “Pubertal changes (Tanner stages).” Information was sufficient to answer most clinical questions. However, UpToDate did not provide a clinically useful graphic listing sentinel events (Tanner stages, peak height velocity, menarche) in typical chronological order with anticipated normal age ranges.

Digital design problems

UpToDate fails to take full advantage of the electronic environment. Inexplicably, in the Contents menu screens, Find in Text is not active, and users cannot navigate using the Page-Up/Page-Down or (Ctrl+)Home/End buttons on their keypads. For example, following Contents through “Patient Education” to “Women’s Health Issues,” users find 48 topics in 19 sections as they try to locate the one of interest.

“Mammography” is integrated into “Screening for breast cancer,” and is not a separate keyword for users who are visually scanning for that word (the only option).

Right-clicking is not supported.

An Editmenu is absent. Windows’ Cut and Paste keyboard conventions (Ctrl-c/Ctrl-v) are supported in the main UpToDate window—but not in other windows.

Bookmarking—the electronic equivalent of the folded page corner—is unavailable, as is a favoriteslist.

Highlighting—electronic yellow, blue, and pink markers—is not supported.

No facility for user-added annotation is available.

Opening multiple windows of information simultaneously is not possible.

Standard resizing of the main UpToDate window (by “grabbing” edges of the window) is not supported.

Finally, an ability to limit searches from the main screen to patient handouts or Lexi-Comp would be desirable.

Pocket version

UpToDate also provides a Pocket PC version. The Pocket PC version searches the same database with a search engine similar to the other versions. It lacks the search history offered in the CD version. Although faster than searching through a textbook, an average UpToDate search on Dell Axim X5 400 mHz (Intel PXA250) with 64 MB RAM and a SanDisk 1G CompactFlash card took about 10 to 27 seconds, plus more time to navigate through the long and sometimes cumbersome submenus.

The primary intent for palmtop medical software is point-of-care use. The extensiveness of UpToDate’s text—occupying 800 MB of the 1G card—and limited speed of palmtop computers often makes this impractical. However, it should be emphasized that this rich, inclusive database is a valuable reference when time is not at a premium. Certainly, the Pocket PC version provides ultimate portability. Users contemplating use of UpToDate/Pocket PC should budget for a highend Pocket PC and a 1G card.

Conclusions

UpToDate is a clinically useful, searchable database of medical information updated every 4 months and available on CD, online, and for souped-up Pocket PC palmtop devices. It is almost as if UpToDate purposely incorporated an unsophisticated search engine and did not optimize the electronic environment to flaunt its forte—the clinical database.

For family physicians, the database is of sufficiently high quality, and it will be even more useful as the pediatric module expands. We hope that future enhancements to UpToDate will include clinical bullets for point-of-care use and optimization of the electronic environment. The latter should include implementation of Boolean searches; bookmarks; highlighters; a user-added note facility; copying and pasting of references; links to drug cost information; and—the whole point of Windows! — multi-window capability and conventional window resizing.

Patient information needs to be more accessible to be clinically useful. Also, versions of the patient information handouts with less complex language to appeal to the average patient would be helpful. As is, UpToDate’s importance will only increase. Highly recommended.

Correspondence
Gary N. Fox, MD, Mercy Health Partners Family Practice Residency Program, 2200 Jefferson Ave, Toledo OH 43624-1117. E-mail: foxgary@yahoo.com.