Hos Assignment 12th Night



Pressure ulcers are insidious complications that affect approximately 2.5 million patients and account for approximately US$11 billion in annual health care spending each year. To date we are unaware of any study that has used a wearable patient sensor to quantify patient movement and positioning in an effort to assess whether adherence to optimal patient turning results in a reduction in pressure ulcer occurrence.


This study is a single-site, open-label, two-arm, randomized controlled trial that will enroll 1812 patients from two intensive care units. All subjects will be randomly assigned, with the aid of a computer-generated schedule, to either a standard care group (control) or an optimal pressure ulcer-preventative care group (treatment). Optimal pressure ulcer prevention is defined as regular turning every 2 h with at least 15 min of tissue decompression. All subjects will receive a wearable patient sensor (Leaf Healthcare, Inc., Pleasanton, CA, USA) that will detect patient movement and positioning. This information is relayed through a proprietary mesh network to a central server for display on a user-interface to assist with nursing care. This information is used to guide preventative care practices for those within the treatment group. Patients will be monitored throughout their admission in the intensive care unit.


We plan to conduct a randomized control trial, which to our knowledge is the first of its kind to use a wearable patient sensor to quantify and establish optimal preventative care practices, in an attempt to determine whether this is effective in reducing hospital-acquired pressure ulcers.


Pressure ulcers are insidious complications that affect approximately 2.5 million patients and account for approximately US$11 billion in annual health care spending each year [1]. Acutely ill patients are at risk for the development of pressure ulcers due to immobility, reduced perfusion, and prolonged duration of mechanical ventilation [2]. In 2008 the Centers for Medicare and Medicaid Services discontinued reimbursement for facility-acquired pressure ulcers (also known as hospital-acquired pressure ulcers, HAPUs), as these are considered an avoidable complication often described as a “never event.” Given the tremendous burden that pressure ulcers place on the individual patient and the health care system, there is a substantial need for improved prevention methods [3, 4].

Pressure ulcers form when there is sustained pressure, predominately over bony prominences such as the sacrum, heels, occiput, and shoulders. Unrelieved pressure causes compression of cellular tissue, impaired blood flow, and can lead to localized tissue damage and cellular death. Pressure ulcers initially appear as areas of reddened skin but can quickly develop into large open wounds if the pressure is not relieved.

To prevent pressure ulcers, the currently accepted standard of care is to turn patients at least every 2 h, day and night. However, there are no published research studies that support the every 2-h turning schedule in critically ill patients. Notwithstanding, studies have estimated that compliance with patient-turning protocols are around 60 % and that a significant number of patients are not being turned as frequently as recommended [5, 6]. In intensive care units (ICUs), compliance to turning protocols is even lower, ranging from 38 to 51 % [7, 8]. Potential explanations for this low compliance include low prioritization of turning, difficulty in monitoring a patient’s position, ineffective turn reminders/alerts, and sub-optimal caregiver staffing ratios – all of which hinder efforts to prevent pressure ulcers.

To improve adherence to turning as a preventative practice, Leaf Healthcare Inc. (Pleasanton, CA, USA) has developed a patient-monitoring system designed to optimize patient-turning practices. This system is composed of a small, single, wearable patient sensor that adheres to a patient’s chest, similar to a standard telemetry electrode. The sensor communicates wirelessly to a central monitoring station about the patient’s current position (upright, supine/back, right side, left side) and time-to-next-turn. Data for all patients are displayed on a User-Dashboard, allowing staff to easily identify patients who are in need of turning. The system also allows caregivers to identify restricted positions due to existing pressure ulcers or surgical wounds, as well as to create personalized turning protocols, by varying the degree of turn angle and/or tissue decompression thresholds.

Two single-center clinical studies have been completed using the monitoring equipment. A small first-in-human Institutional Review Board (IRB)-approved study was performed to test the feasibility of the patient sensor, mesh network, and monitoring system, and to collect baseline data regarding turning protocol compliance. After this, a second single-center IRB-approved clinical study tested the efficacy, usability, and safety of the monitoring system (ClinicalTrials.gov #NCT02005692). After implementation of the system, compliance to patient turning was reported to have increased significantly from 64 to 98 % [9].

To date, the study team are unaware of any study that has used a wearable patient sensor to quantify patient movement and positioning, in an effort to assess whether adherence to optimal patient turning results in a reduction in pressure ulcer occurrence. As such, the following protocol describes the study to be conducted to evaluate whether optimal patient turning, defined as regular turning every 2 h with at least 15 min of tissue decompression, reduces HAPUs in acutely ill patients.



This study is a single-site, open-label, two-arm, randomized controlled trial. As this is an open-label trial, a short observation pilot study will be undertaken during installation and testing of the patient-monitoring system to account for potential observer bias (the Hawthorne effect). The true intent of the technology will be obfuscated to patients and clinical staff in an effort to record baseline data that more accurately represent current turning practices within these units. These pilot data will also be used to determine minimum turning thresholds for the main study and to assess the representativeness of the main study’s control group. A convenient sample of 25 subjects will be enrolled from each participating unit.


All subjects will be randomly assigned, with the aid of a computer-generated schedule, to receive either standard care (control group) or optimal pressure ulcer preventative care (treatment group). To minimize the risk of predicting the treatment assignment, randomization is performed in permuted blocks of two, four, and six, with random variation of block sizes. To minimize bias at the patient level, randomization is further stratified by unit of admission and admitting service, either medicine or surgery (Fig. 1). Once randomized, each subject will receive a nominal study identification number based on the unit of admission and admitting service. The lowest available number will be provided to each subject in sequential order.

Fig. 1

Randomization schema. Stratification by unit and admitting service team

Study population

This study will enroll all patients admitted to two ICUs. These patients are critically ill with exacerbations of acute and chronic medical conditions, or are receiving aggressive post-surgical care for neurological, cardiac, or trauma-related conditions. They typically have altered levels of consciousness and are dependent on clinical staff for activities of daily living. Due to immobility and other clinical factors these patients are at high risk for developing pressure ulcers. As it is infeasible to gain written consent from this patient population and, due to the minimal risk of the study procedures, a waiver of individual authorization has been granted by Stanford University’s IRB (see Ethics approval section). Therefore, all patients over the age of 18 years admitted to one of the two ICUs will be enrolled in the study. Patients under 18 years of age, those with a known allergy to skin adhesive, or who possess a physical barrier preventing the application of the monitoring sensor, are not eligible for inclusion.


Upon admission to a study unit and as part of standard care, the patient receives a complete “head-to-toe” assessment of their skin. This is performed by two registered nurses (RNs). Any pre-existing pressure ulcer that is detected during this assessment is documented in the electronic medical record (EMR). After initial standard admission procedures are performed, and within the first hour of arrival, nursing staff will place a patient sensor (Leaf Healthcare, Pleasanton, CA, USA) in a predefined location on the patient’s chest (Fig. 2). If an exclusion condition is present, this will be documented and the patient will not be enrolled in the study and will not receive a patient sensor.

Fig. 2

Location of patient-monitoring sensor. *Used with permission from Leaf Healthcare, Inc.

For patients included in the study, the unit secretary will provide nursing staff with a patient sensor after associating the sensor’s serial number with the patient’s specific information. This is automatically derived from an Admission-Discharge-Transfer (ADT) data stream communicating directly with the Leaf Patient Monitoring System. At this time the unit secretary will open an envelope, pre-filled with the computer-generated randomized group allocation. A patient sticker will be affixed to the back of the randomization card and stored in a secure location for later retrieval and verification of correct randomization and enrollment. The patient will then be enrolled in either a treatment or control group.

When a subject is enrolled in the treatment group, the User-Dashboard will be turned “on,” allowing the sensor to communicate the patient’s position and movements (Fig. 3). The clinical team will review this information and be guided by visual displays to provide pressure ulcer-prevention turning practices. With the use of the User-Dashboard, nurses will be prompted to perform pressure ulcer preventative care, namely patient turning with satisfactory tissue decompression for 15 min, at least every 2 h (optimal). If a patient does not receive the full time of tissue decompression, the Leaf Patient Monitoring System automatically adjusts, proportionally reducing the time-to-next-turn. For example, if the patient was on their back and moved to their right side, but returned to their back within 7–8 min, the time-to-next-turn will be adjusted reciprocally from 2 h to 1 h. This ensures that patients receive at least 15 min of tissue decompression every 2 h.

Fig. 3

User-Dashboard – Leaf Patient Monitoring Interface. *Used with permission from Leaf Healthcare, Inc.

If the subject is enrolled in the control group, the unit secretary will turn the User-Dashboard “off” by selecting the “control patient” check-box in the Leaf Patient Monitoring System. This will turn the sensor information off and therefore not display any information to the User-Dashboard. Patients in the control group will continue to receive standard care practices, that is, pressure ulcer-prevention activities initiated by nurses using their usual care routines. Nursing care will not be “optimized” by the Leaf Patient Monitoring System, with the system withholding any data related to turning frequency, position schedule, and decompression time. The patient sensor will continue to collect these data for the purposes of research analysis only.

Each patient’s participation within the study begins within the first hour of arrival to a study unit and ends upon discharge or transfer from the study unit. Upon leaving the unit, the patient sensor will be removed and the patient will be discharged automatically from the Leaf Patient Monitoring System.


The Leaf Patient Monitoring System is a proprietary system developed by Leaf Healthcare Inc. (Pleasanton, CA, USA) [10]. The Leaf Patient Monitoring System is a wireless monitoring system that enables personalized turning protocols. A wireless, wearable, single-patient sensor communicates a patient’s body position and movement, through a proprietary mesh network of relay antennas placed throughout the study units to a central monitoring system (Fig. 4). The Leaf Turn Management System accesses an industry standard Health Level Seven International (HL7)-ADT data stream. This allows for simplified and accurate patient enrollment and seamless tracking of patient movement throughout the study units. For data security, all data generated from the Leaf Patient Monitoring System are transferred and stored within the institution’s firewall to secure servers. The Leaf Turn Management System is the User-Dashboard that displays an individual patient’s positioning and movement data to the clinical team.

Fig. 4

Wearable patient sensor communication network. *Used with permission from Leaf Healthcare, Inc.

Data management

Proprietary turning data will be acquired from the Leaf Patient Monitoring System. Compliance to preventative turning care will be defined by two measures. The first will sum the overall time the patient is overdue for turning care, divided by the total monitoring time, and will be reported as a percentile of time. For example, if a patient was on the unit for 4 h and did not receive preventative turning care at the second hour, but instead at the third hour (1 h overdue), then the patient’s care was in compliance for 3 of the 4 h of their length of stay. This would, therefore, represent a time in compliance of 75 %. If they received the turn on schedule, the compliance would be at 100 %. The second measure will divide a patient’s length of stay into 2 h turning blocks. In the above example, if the patient was turned once in their 4 h stay, compliance to turning would equal 50 %. Although we believe that the time in compliance is more appropriate than the proportion of turns within 2 h blocks, we will test and report on both measures as the latter reflects current clinical care practices.

Pressure ulcer staging

The National Pressure Ulcer Advisory Panel (NPUAP) staging criteria will be used to stage all HAPUs [11]. Staging will be completed by an expert RN blinded to the patients’ study allocation. Adjudication will occur with the study team for any wounds that are difficult to stage.

  • Stage I, non-blanchable erythema: intact skin with non-blanchable redness of a localized area usually over a bony prominence. Darkly pigmented skin may not have visible blanching; its color may differ from the surrounding area. The area may be painful, firm, soft, warmer or cooler as compared to adjacent tissue. Category I may be difficult to detect in individuals with dark skin tones. This stage may indicate “at risk” persons

  • Stage II, partial thickness skin loss: partial-thickness loss of dermis presenting as a shallow, open ulcer with a red-pink wound bed, without slough. This may also present as an intact or open/ruptured serum-filled or sero-sanginous-filled blister. It presents as a shiny or dry, shallow ulcer without slough or bruising*. This category should not be used to describe skin tears, tape burns, incontinence-associated dermatitis, maceration or excoriation. *Bruising indicates deep tissue injury

  • Stage III, full-thickness skin loss: full-thickness tissue loss. Subcutaneous fat may be visible but bone, tendon or muscle are not exposed. Slough may be present but does not obscure the depth of tissue loss. This may include undermining and tunneling. The depth of a category/stage III pressure ulcer varies by anatomical location. The bridge of the nose, ear, occiput and malleoli have no (adipose) subcutaneous tissue and category/stage III ulcers can be shallow. In contrast, areas of significant adiposity can develop extremely deep category/stage III pressure ulcers. Bone/tendon is not visible or directly palpable

  • Stage IV, full-thickness tissue loss: full-thickness tissue loss with exposed bone, tendon or muscle. Slough or eschar may be present. This stage often includes undermining and tunneling. The depth of a category/stage IV pressure ulcer varies by anatomical location. The bridge of the nose, ear, occiput and malleoli have no (adipose) subcutaneous tissue and these ulcers can be shallow. Category/stage IV ulcers can extend into muscle and/or supporting structures (e.g., fascia, tendon or joint capsule) making osteomyelitis or osteitis likely to occur. Exposed bone/muscle is visible or directly palpable

Data collection

Patient demographics and clinical data will be obtained from the Stanford Translational Research Integrated Database Environment (STRIDE). STRIDE is a clinical data warehouse that is maintained by the Stanford University School of Medicine and is updated in near real time directly from the EMR used at Stanford Health Care. All data generated from the study will be managed and stored using REDCap electronic data capture tools, hosted by Stanford University. REDCap (Research Electronic Data Capture) is a secure, web-based application designed to support data capture for research studies [12].

Study endpoints


The following endpoints will be evaluated:

  • Difference in compliance rates with preventative turning practices between the treatment and control groups, as continuously measured by the Leaf Patient Monitoring System during the patients’ ICU stay

  • Difference in the proportion of HAPUs between the treatment and control groups


Secondary measures are to:

  • Evaluate and describe, any differences identified in hospital-acquired pressure ulcer rates and stages between medical and surgical patient groups

  • Develop explanatory models of contributing clinical factors to HAPUs for the three groups: all ICU patients, medical patients only, and surgical patients only

Sample size calculation

Using a two-sided Z test of the difference between proportions with 80 % power and a 5 % significance level, a sample size of 1812 patients, 906 in each group, will be sufficient to detect a clinically important difference of 50 % between the groups in the rate of HAPUs. This assumes that a 50 % difference represents a change from a prevalence rate of 5 % for pressure ulcers in the control group, to a rate of 2.5 % for pressure ulcers in the treatment group. Conservatively, with an estimated enrollment of 200 subjects per month, study enrollment is expected to be between 7 and 10 months’ duration.

Statistical analysis

Data analysis will be conducted using IBM SPSS Predictive Analytics Software (v20, Armonk, NY, USA), after data cleaning and organization within Excel 2010 (Microsoft, Seattle, WA, USA). A consultant statistician, external to the study team and trained in quantitative methods will oversee data analyses.

Descriptive analysis

Demographic and clinical characteristics will be summarized for each treatment group and reported. Frequencies will be determined for count data, such as the proportion of patients with pressure ulcers, and descriptive summaries will be used for continuous data, such as percentage of compliance time. Measures of central tendency (mean, standard deviation, median, minimum, and maximum values) will also be reported.

Primary endpoint

An independent-samples Student’s t test will be used to evaluate for differences in preventative turning compliance between the two groups. To test for differences in pressure ulcer rates between the treatment groups, a 2 × 2 table with chi-square statistic will be used. To further evaluate for differences in pressure ulcer rates between treatment groups over time, a Kaplan-Meier survival analysis with log-rank test will be used. For all analyses, tests are deemed to be significant if significance is less than p <0.05. Device-related pressure ulcers from catheters or tubes are generally unrelated to turning and are not included in this analysis.

Secondary endpoint

To test for differences in HAPUs between patients based on admitting service (medical or surgical), a Kruskal-Wallis rank test will be used. Stages of pressure ulcers (based on NPUAP staging criteria) will represent the ranks to be measured. A difference will exist if the significance level of the test statistic is p <0.05.

In creating explanatory models, multivariate logistic regression will be used. First, univariate testing of each variable in relationship to the endpoint (hospital-acquired pressure ulcer) will be performed. Only variables having a maximum significance with the endpoint of p <0.20 will be included in the multivariate model. Age and sex will be reintroduced to the models if they fail to meet the inclusion criteria and all other variables will be entered into the model in a forward stepwise fashion. This will be rerun until only significant variables remain. Three logistic regression models will be constructed: all ICU patients, only medical patients, and only surgical patients.


This study investigates the effect of optimal turning, defined as patient turning every 2 h with at least 15 min of tissue decompression, on reducing HAPUs. It compares outcomes in the treatment group with a control group. The treatment group consists of patients receiving clinical care that is optimized by the Leaf Patient Monitoring System (Leaf Health Care, Pleasanton, CA, USA). This system monitors patient positioning and advises when a patient turn is due, based on preset criteria, and allows for personalized care practices.

The strength of this study is its robust design. A randomized control trial is planned, which to our knowledge is the first of its kind to use a wearable patient sensor to quantify and establish optimal preventative care practices, in an attempt to determine whether these are effective in reducing HAPUs. Prior studies have used an assortment of technical solutions, including environmental and mattress sensors measuring body temperature, estimated body position, and surface compression [13]. However, these studies have consistently failed to assess outcomes: specifically, whether these approaches actually reduce HAPUs. In this study, any positive results will be of immediate benefit as much of the study’s methodologies maintain standard care practices. Therefore, positive findings will be readily translated into clinical practices.

However, there are important challenges in this study. The first challenge is maintaining group assignment. We plan to randomize patients across two ICUs, involving 58 beds and over 300 RNs. Over the course of the ICU length of stay a patient may receive care from many RNs. During the study period it is very likely that an individual nurse will care for patients in both treatment and control groups. Therefore, there is a risk that nurses will transfer and apply knowledge of optimal turning practices between groups. In addition, although nurses remain within the patient’s room for their entire shift, it is plausible that communication regarding prevention care practices occurs during breaks, or with nurses in adjacent patient rooms. To identify and account for potential bias, a small pilot study will be conducted prior to the primary study to establish baseline adherence rates. Twenty-five patients from each unit will receive unmarked sensors from the research team. Nurses will be masked as to the sensors’ true intent and the sensors will collect the same patient-positioning information as the primary study. These data will be used to compare data derived from the control group, to assess for any potential cross-over or observer bias effects.

Another challenge within this study is the number of software steps required to ensure accurate function, feedback and group assignment. The clinical care team receive data from the User-Dashboard about a patient’s position and time-to-next-turn. However, the system requires that the unit secretary accurately enters the correct group assignment. Also, due to the complexity of clinical software applications, accessing the User-Dashboard is a two-step process for staff. Staff must separately open the software application in addition to the EMR. Once activated, the software reduces the technical burden on clinical staff as the User-Dashboard accesses a HL7-ADT feed. This automatically provides complete patient demographic data and tracks patient location throughout their ICU length of stay, minimizing the need for staff to continually update this information.

To further ensure that clinical staff complete study procedures, daily monitoring of study units will occur. To ensure that unit secretaries are correctly assigning patients based on the computer-generated randomization schedule, a patient sticker will be applied to the back of the randomization card and placed in a secured box at the point of enrollment. On a regular schedule these data will be collected, stored, and cross-referenced with the patient allocation entered into the Leaf Patient Monitoring System.

Finally, this study brings together research personnel from both clinical and technical teams of a major academic health system, working together to implement a large clinical trial within a complex care environment. As organizations focus on improving clinical care environments, building efficiencies in clinical research and in adopting and testing innovative technologies is imperative in improving health care delivery and clinical outcomes for patients.

Trial status

At time of submission this trial is open to enrollment.

Acknowledgements and Funding

Authors would like to acknowledge the support of Nancy Lee, VP Patient Care Services and Dr. Pravene Nath, Chief Information Officer, Stanford Health Care. We also acknowledge the direction and support of Dr. Chris Furmanski, Director of Innovation, Rajiv Ramdeo (Informational Technology), the staff and management of the E2 and E29 clinical ICUs, including unit managers, Maureen Fay and Kathy Seppala, as well as Chris Nguyen, Alex Kreis, Michael Pihulic, Annemari Cooley, and Mark Smith from Leaf Healthcare, Inc. This study is co-sponsored by Stanford Health Care and Leaf Healthcare, Inc.


HAPUHospital-Acquired Pressure Ulcer
HL7Health Level Seven International
ICUIntensive Care Unit
PUPressure Ulcer
EMRElectronic Medical Record
RNRegistered Nurse


Competing interests

Leaf Healthcare, Inc. was co-founded by Dr. Barrett Larson MD, a physician employed at Stanford University. Leaf Healthcare, Inc. is a co-sponsor of this study and provides monitoring hardware, proprietary software, and technical support.

Authors’ contributions

All authors contributed to the development and design of the protocol (DP, BB, KE, JS, ML, and BM). DP, BB, and KE developed the analysis plan and applied for institutional support. DP has drafted the manuscript with critical input from all other authors who have read and approved the final manuscript (BB, KE, JS, ML, and BM).


1. National Pressure Ulcer Advisory Panel, European Pressure Ulcer Advisory Panel, Pan Pacific Pressure Injury Alliance . In: Prevention and treatment of pressure ulcers: clinical practice guideline. Haesler E, editor. Perth: Cambridge Media; 2014.

2. Cooper KL. Evidence-based prevention of pressure ulcers in the intensive care unit. Crit Care Nurse. 2013;33(6):57–66. doi: 10.4037/ccn2013985.[PubMed][Cross Ref]

3. AHRQ. Preventing pressure ulcers in hospitals: a toolkit for improving quality of care. Rockville, MD: Agency for Healthcare Research and Quality; 2014.

4. Vangilder C, MacFarlane G, Meyer S. Results of nine international pressure ulcer prevention surveys: 1998–2005. Ostomy Wound Manag. 2008;54(2):40–54.[PubMed]

5. Gunningberg L. Are patients with or at risk of pressure ulcers allocated appropriate prevention measures? Int J Nurs Pract. 2005;11(2):58–67. doi: 10.1111/j.1440-172X.2005.00503.x.[PubMed][Cross Ref]

6. Lyder CH, Preston J, Grady JN, et al. Quality of care for hospitalized medicare patients at risk for pressure ulcers. Arch Intern Med. 2001;161(12):1549–54. doi: 10.1001/archinte.161.12.1549.[PubMed][Cross Ref]

7. Krishnagopalan S, Johnson EW, Low LL, et al. Body positioning of intensive care patients: clinical practice versus standards*. Crit Care Med. 2002;30(11):2588–92. doi: 10.1097/00003246-200211000-00031.[PubMed][Cross Ref]

8. Schallom L, Metheny NA, Stewart J, et al. Effect of frequency of manual turning on pneumonia. Am J Crit Care. 2005;14(6):476–8.[PubMed]

9. Tarver C, Schutt S, Pezzani M. We’re sensing you! A multi-phase clinical trial examining innovative technology to improve patient-turning compliance. Paper presented at the ANCC National Magnet Conference. Dallas, Texas. 2014.

10. Leaf Healthcare, Inc. Solution overview: the next step in wireless patient monitoring: Leaf Healthcare Inc., 2014. accessed February 12th, http://leafhealthcare.com/pdfs/LH_SolutionOverview_1560AA_060214.pdf

11. National Pressure Ulcer Advisory Panel Pressure Ulcer Stages. Secondary NPUAP pressure ulcer stages.

12. Harris P, Taylor R, Thielke R, et al. Research Electronic Data Capture (REDCap) – A meta-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. doi: 10.1016/j.jbi.2008.08.010.[PMC free article][PubMed][Cross Ref]

13. Marchione FG, Araújo LMQ, Araújo LV. Approaches that use software to support the prevention of pressure ulcer: a systematic review. Int J Med Inform. 2015;84(1):725–36. doi: 10.1016/j.ijmedinf.2015.05.013.[PubMed][Cross Ref]

Mark Purdy, who has been a sports columnist for this newspaper since 1984, will retire from that job in August. He is writing a series of columns recounting the most memorable elements of his four decades as a sports journalist. This time: Best and worst assignments.

For a sportswriter, that first big assignment is really something. Your first plane flight to cover a game? So exciting. My first was to a football game in Pittsburgh. And my bosses paid for it! I felt as if I had arrived in the business.

Related Articles

Hundreds more flights followed. Gradually, they became less exciting. Well, one still was–and ironically, it was another trip to Pittsburgh. On approach, the landing grear failed to drop. The captain walked down the aisle, peeled up the carpet to crawl inside the fuselage, then manually cranked down the wheels. He returned to the cockpit and warned us to prepare for a possible bad outcome in case the wheels didn’t hold. Emergency crews lined the runway. We braced ourselves.

At this point, my journalist companion on the flight, a witty pro football scribe named Ray Buck, turned to me and said: “You know, Purd, if this plane goes down, they’ll list the victims alphabetically. So my name will appear before yours in the obituary.”

That’s sportswriter humor for you. I forced a laugh. We landed . . . gently, safely and not alphabetically. We left by the emergency exit, took a taxi to the stadium and covered the game. Another day on the job. Assignments are assignments. I’ve had some great ones and lousy ones. You go where they send you. I never complained. I always found interesting stories.

Once, the big boxing matches were my favorite trips.You can ask anyone who’s been to a major fight in Las Vegas. The mojo in the days leading up to the event is unmatched, with celebrities wandering around — hey, is that Dustin Hoffman entering the elevator? — and the sharpies flying in from both coasts to lay down cash at the betting windows. Pinky rings proliferate. Smarmy perfume odors ooze. By the night of the main event, when spotlights hit the ring and only the two fighters and referee are left inside to settle business  . . . it’s just hard to match.



Those moments used to happen frequently, back when fighters were household names. Muhammad Ali, Larry Holmes, Roberto Duran, Marvin Hagler, Thomas Hearns, Evander Holyfield, Aaron Pryor . . . all rich personalites. But no boxer was a more arresting column subject than Mike Tyson. When he dominated the heavyweight division, Tyson was a must-see magnet. Chaos reigned in his entourage. One hotel clerk told me that when housekeeping crews cleaned the rooms on Tyson’s floor, they were ordered to pull their carts inside rooms while doing their work. If left in the hallway, the carts would be defoliated.

After one Vegas fight, I wound up at a hotel lounge with Hunter S. Thompson, the gonzo writer. A mutual friend introduced us. Many drinks followed, as well as an extended conversation that touched on the “Ben Hur” film soundtrack, leeches and the NFL. I’ll leave it at that. As the night wore on, singer Wayne Newton and boxer Leon Spinks passed through the bar. Spinks’ had a bizarre-looking bodyguard named Laurence Turead. He would later be known as “Mr. T.”  What a scene. More drinks were consumed. The next morning, I was extremely hung over.

In retrospect, I guess that night would qualify as both a “best” and “worst” work experience.

Yet incredibly, as I went over the last four decades to select my five best and worst assignments, I found myself leaving all boxing events off either list. There was too much other stuff. Both ways. But we’ll start with my least delightful assignments. Here they were:


This was the shortest distance I ever traveled to cover a story. It’s also a story I wished I’d never had to cover. Tillman, the San Jose native and Arizona Cardinal safety who left the NFL to join the Army Rangers after 9/11, perished on duty in Afghanistan in the spring of 2004. He was a victim of friendly fire under circumstances that remain up for debate because of the government’s botched handling of his case.

Tillman’s public memorial service was held at the San Jose Municipal Rose Garden, within blocks of my house. So I strolled over with my notebook to try and capture the most raw emotional event I’ve witnessed. Around 2,000 people attended. Fifteen satellite trucks were parked on the street. Radio host Jim Rome emceed. Sen. John McCain spoke. Then the people who truly knew Tillman stood up to speak. It was riveting.

“Thank you for coming,” said his brother Richard, holding a Guinness beer in his hand. “But with all respect to those who have been up here before me, Pat’s not with God. He’s not religious. He’s dead. It was amazing to be his little brother.”

“If Pat were here today,” said Paul Ugenti, a high school friend, “he would be saying, ‘Get off your ass and get hold of yourselves.’ ”

Jake Plummer, Tillman’s teammate in Arizona, gave the most touching remarks when he talked about the philosophical discussions the two men had.

“You’d walk away from one of those sessions and say to yourself, ‘Man, I’ve got to become more of a thinker.’ ” said Plummer. “To me, the saddest part of all is to not know what he planned next.”

My personal belief is that had he lived, Tillman would have settled in San Jose after his football retirement and either run for office or started up a company. After the memorial service that day, I walked home and wrote: “Tillman’s 10th high school reunion would have been this summer. It will now happen without him. Guinness will be consumed. Stories will be told about an honest man who would have been proud of what his honest friends said Monday.”


Do I really need to explain? The NFL has come to dominate our athletic culture so completely–and admittedly, does provide its share of thrilling moments–that we tend to forget how the league also produces two of the least compelling regularly scheduled concepts known to modern spectator sports: Preseason exhibition games and the annual college draft.

Yes, it’s a crime that the 49ers and Raiders and every other team in the league forces season ticket holders to purchase seats for games that mean nothing. But at least there’s some football action involved. You can’t say the same for the draft, which has grown from a one-day event into a three-night prime time production where people . . . stand up and read off names.  That’s it. Despite all the bells and whistles, that’s it.

I understand how the ramifications of draft night can be significant. I understand that intrigue is involved with trades and moves. But the bottom line is,  you never truly know how a rookie draft pick will do until he is on the field competing with NFL players. Analyzing the success of a team’s draft five minutes after it ends is a fool’s errand. You’re throwing darts.  It takes at least two or three years before you can draw real conclusions. The draft happens in late April. Every year, I pulled for the Warriors and Sharks to go far enough in the playoffs so that I could avoid covering the name-reading. Thank you, Warriors and Sharks, for recently never letting me down.


You may be surprised to find this hallowed event on my “worst assignment” list. But the answer lies in a question many people never ask themselves: “Gee, did you ever notice something different about Masters telecasts compared to every other major golf championship? There are no overhead blimp shots. Why is that?”

I can tell you why. The poobahs that run Augusta National don’t want you to know that across the street from their cherished course are a Circle K convenience store and Jay’s Music Super Center. Just down the road are four or five tattoo parlors.  The simple truth is that while inside Augusta’s gates is one of the world’s most beautiful golf courses, outside the gates is a city that (sorry, Augustans) is a distinctly non-glamorous city. It mostly serves a nearby army base and is not exactly stocked with first class hotel rooms. Once, making a late reservation, I wound up in a place where I’m fairly certain that heroin was being shot up down the hall.

Also, I never felt quite comfortable with the Masters’ elitist nature. My preferred golf tournaments are the US Open and British Open, in which anyone on earth can attempt to earn a spot through a series of qualifying events. The Masters’ elders set up a limited field to their liking through a series of  “standards” that often wind up excluding great up-and-coming players. This gives the Masters the weakest field of the four major championships but almost always provides the tournament with a well-known winner.

The first Masters I covered, in 1975, was also the first Masters in which an African-American (Lee Elder) was invited to play. That was 11 years after the congressional Civil Rights Act. Lee Trevino used to change his shoes in the parking lot instead of the locker room because he felt uncomfortable in the clubhouse. I assume that vibe has changed, now that the Augusta members have begun to admit women and minorities. I wouldn’t know. About 20 years ago, I began declining the Masters assignment from my bosses. I knew we should send people who were far more excited to go than I was.


This, too, may be a surprise entry. My passion for postseason baseball and hockey is well known–but that’s as a viewer or a paying customer. For a writer on deadline, nothing is worse than a baseball playoff game that goes into extra innings or a hockey game that goes into sudden-death overtime.

Football and basketball can go into extra periods, as well. But it happens very infrequently in the NFL. And in NBA games, there’s so much scoring that a multi-overtime game is a rarity. In baseball, with no clock, a game could theoretically last forever. And it’s always a long slog through each inning with the deadline clock ticking. In hockey, there’s a different problem. The sudden-death component means the game could end instantly, which means writers must simultaneously compose two stories (one if the home team wins, one if it loses) and possibly three (if the desk needs a story that says “at press time, the game was still not decided.”).

This explains why, in the late innings or extra periods with writers pounding away, the most common phrase heard in the press box after a crowd roar is: “What happened?”  We scribes look up from our keyboards and hope there’s a decent replay. Does anyone recall that 2014 playoff game between the Giants and Nationals in Washington D.C. that went 18 innings and lasted six hours and 23 minutes? It’s still the longest postseason baseball game in history. It was an open press box.  As the night wore on, the temperature dropped into the low 50s or high 40s. I hadn’t brought a heavy jacket. I tried to type with gloves. Couldn’t. The game finally ended well past midnight after a Brandon Belt home run. I wound up writing my column about unheralded Giants’ reliever Yusmeiro Petit. He sat in the chilly bullpen for 11 innings and admitted his fingers were “numb” when he was called upon in the bottom of the 12th.

“The first hitter, I wasn’t feeling the ball,” said Petit, who would save the Giants’ bacon by going six innings and allowing just one hit. Great story. I had about 10 minutes to write it. Against all odds, it was in complete sentences and in English.

The Sharks’ equivalent of that game occurred in 2008 in their own longest game ever, a four-overtime playoff defeat in Dallas. The beloved Los Tiburones lost when defenseman Brian Campbell, on his 68th shift of the evening, committed a penalty and gave the Stars a power play which they converted. It ended the series. In a silent postgame dressing room, a drained Campbell said: “There’s not much I can feel about it.” I rushed upstairs to get that quote in my final edition column, which was barely comprehensible. Another perfect example of how I could love and hate my job in equal measure.


Wow, did I create a ruckus in January 2004 when the Super Bowl was played in Houston for the first time in more than 30 years. The lead of my first column that week was: “Yikes. Get me out of here.”

No, that wasn’t subtle. But everyone who travels has favorite road cities and non-favorite road cities. Houston was my most non-favorite. Lots of people live there, so the city must be a decent place to raise a family. For a business trip, however, Houston has terrible logistics. The city’s major airport is a long hike from downtown. The downtown hotels are far away from the football stadium. And driving between any of these places, the freeway traffic is horrible. In the summer, the weather is oppressively humid. In the winter, it can be rainy and windy.

As I opined: “Every time someone says San Jose is spread out and haphazardly planned, I tell them to visit Houston. There are no real zoning laws. You can motor down a street and pass a barbecue joint, followed by a church, followed by a nice apartment complex, followed by a strip club, followed by a public library, followed by a tattoo parlor. It’s like driving through a city designed by the stoner teens in “That ’70s Show.”

When local residents read these words, they were not amused. I wound up having to go on local television to defend myself, not entirely successfully. On my latest trip to Houston, for the Raiders’ playoff game last January, I noted that the downtown is a little more lively. I still plan no vacations there in retirement.

Now, here is the happier portion of the program with my five all-time most favorite assignments:





More than once over the years, I have told the tale of how I wound up coming face-to-face with Fidel Castro at a bowling alley in Havana. The details were so preposterous, you couldn’t make them up.

The 1991 Pan Am Games — basically the western hemisphere Olympics — were a chance for Cuba to show off itself to the world as the event’s host. Except the country was in a profound recession/depression. The sports venues were not first rate with the exception of a brand new bowling alley that had been built for the Games. Castro was famous for not announcing his daily schedule in advance (so that no one could be in waiting to harm him) and one day he decided to take in the bowling competition. He wanted the best seats, which happened to be in the press row.

As applause broke out at Fidel’s entrance, we were ordered to move. As I made my way to the end of the aisle, there he was.  I made meaningful eye contact as he passed me and plopped down in my exact spot. That became the lead of my column: “Fidel took my seat.” With a Havana dateline, I figured that would get people to the second paragraph.

The Castro encounter was hardly my only Havana adventure. This was in the era when Cuba was officially closed to Americans, so it was a fascinating time to be there. As journalists, we were trailed almost everywhere we went by “security people” on scooters. But one day, myself and a friend from the Los Angeles Times rented bicycles and purposely headed the wrong way down one-way streets to successfully evade our followers and talk to some average Cubans without monitors nearby. We also tracked down the home where A’s slugger Jose Canseco and his brother lived as toddlers before their parents took them to America.  Across the street was a woman who was married to Jose’s cousin.

“Yes, I remember Jose and his twin brother, Osvaldo,” she told us. “They were cute. I hear Jose is very famous now. That’s great. I’m glad, because he has done me no harm. But we haven’t been interested in him and he hasn’t been interested in us. There’s been no communication between us.”

Her words made me sad. I’m not sure if, with the new USA-Cuba diplomatic initiatives, the Canseco family has reconnected. I’m hoping so.




The Rose Bowl is my favorite annual sports event.  New Year’s Day in Pasadena is a sublime sports experience, aesthetically and competitively. But as a columnist, it was not my favorite event to cover. That would be the Super Bowl, which is the country’s most-watched sporting event and also the craziest to cover, thanks to the sponsorship circus it has become and the relentless hype drum pounded by the NFL.  I knew that whatever I wrote would draw readers’ interest.

Thus, I doggedly plowed through Media Day, where cable television “reporters” from MTV or the SyFy channel wearing costumes vie to throw out the first stupid questions of the week and always succeed. Early on, the zoolike atmosphere frustrated me. Then I realized that covering the Super Bowl was really not about journalism. It was a business transaction between the NFL (which used football to sell advertisers’ products) and those of us in the media (which used stories about the Super Bowl to sell advertisements on their broadcasts or websites or newspaper pages). Once I accepted that fact, I could relax and roll with it by finding off-the-beaten path angles that didn’t involve interviewing a nose tackle.

One example: Before Super Bowl XXII in San Diego, I visited Tijuana to talk with a bullfighter there who had also played some football wide receiver at his high school. He told me that football was far tougher than bullfighting because (A) there’s only one bull as opposed to 11 tacklers and (B) bulls never blindside you.

Another example: At Super Bowl XXXV in Tampa, renowned for having the most strip clubs per capita of any American city,  I decided to visit one and get some predictions from the dancers. (They also told me which players had visited their club.) The matchup that year was Ravens vs. Giants.  A performer named Dawn told me she was sure the Giants would win.

“But that’s because I’m anti-Trent Dilfer,” she said, referring to the Ravens’ quarterback. “We trained him here when he was with the Buccaneers and then he goes off and takes another team to the Super Bowl.”

Nothing is more bitter than a bitter stripper football analyst. And of course, Dawn was nakedly wrong. Dilfer and the Ravens did win. I eagerly covered more than 30 other Super Bowls, none more enjoyable than Super Bowl 50 in Santa Clara. That day, I put down the top on my convertible and took surface streets to Levi’s Stadium with absolutely no traffic issues or tie-ups. That’s a story my grandchildren will never believe.




I pulled off the best scam of all time in 1987 when I convinced my newspaper’s bean counters that I should be sent to Perth for the America’s Cup yacht races. Then and now, there aren’t a ton of yacht racing fans in the Bay Area.

However, my pitch was this: At a slow time of year (February), our readers would glom onto the brave story about USA skipper Dennis Conner’s attempt to wrest back the America’s Cup from the Australians who had defeated Conner four years earlier–the first-ever defeat for the United States in the Cup competition.

My ulterior motive, of course, was to get to Australia. I’d always been fascinated by the country and wanted to see it in person. Much to my shock, the bosses bought my pitch. And I was off to the West Coast of Oz. It didn’t disappoint. Neither did the event. Because of the time difference, my daily deadline was 1 p.m., so after covering each race in the morning, I filed my copy and could spend the rest of the day drinking beer and/or going to the beach.


Leave a Reply

Your email address will not be published. Required fields are marked *