Leading Change

Ten years ago this week, I became CIO of CareGroup. On that day, I learned an important lesson about leading change.

Just hours after getting the job I decided that we'd embrace a service oriented architecture (SOA), standardize all desktop/server/storage infrastructure, and implement centrally managed applications. At 8am the next morning, I was scheduled to meet with my 300 staff members and share with them my vision for the future.

Luckily, experienced leaders counseled me on that first day. I discussed my vision with three Board members - Warren McFarlan (Professor at Harvard Business School), John Keane Sr. (CEO of Keane Inc.) and Sam Fleming (CEO of DRI/McGraw Hill). They told me that announcing a strategic plan without engaging all the stakeholders in the process would lead to mixed support and adoption.

Instead of arriving at the 8am meeting with all the answers, I arrived with questions. I explained to the staff that we wanted to improve customer service, encourage innovation, and ensure our work was aligned with the needs of our stakeholders. I challenged them to tell me what they thought we should do. In the first 30 days of my CIO tenure, I met with every staff member in IS as well as every senior manager in CareGroup to gather their priority lists, synthesize their input, and ensure they had a voice in the future. The result was a new IS operating plan focused on getting the basics done right. We clearly communicated the work to be done, the organizational structure which supported that work and the right people to staff the structure. The next step was to implement the changes.

I've long been a fan of John Kotter and his work on leading change in organizations. His broad recommendations to effect change include:

a. Defrost the status quo
b. Take actions that bring about change
c. Anchor the changes in the corporate culture

The planning meetings described above defrosted the status quo. The actions I took to bring about change included:

Create a Vision for Change - the community came together with a vision of a web-centric organization and I broadly communicated it.

Establish a Sense of Urgency - everyone recognized that IT innovation was essential to coordinate clinical care, improve safety and enhance our competitiveness, especially after the merger of Beth Israel and Deaconess

Elicit Executive and Peer Sponsorship - The CEO declared that medication safety, personal health records, and enhanced communication to all levels of the organization were the strategic goals of the entire organization for that year

Communicate Vision to Implement Change - we established steering committees, project charters, project plans, and communication plans

Empower Employees to Implement Change - we aligned responsibility, accountability and authority throughout the IT organization so that managers had the resources and authority they needed to support our improvement efforts. We created a Special Projects team to coordinate the improvement projects without disrupting day to day operations.

Establish Short-term Goals - we created the first web application in healthcare to share data (with patient consent) among multiple organizations. We created the first web-based provider order entry system, and we created the first personal health record to share all hospital data with patients

Encourage Additional Changes - we created a non-punitive culture in which everyone was encouraged to identify mistakes and opportunities for process improvement

Reinforce Changes Made as Permanent - we built standard processes to deliver service, prioritize new projects, and communicate our multi-year plans to the community

That first year, we implemented strong project management methodologies, eliminated unnecessary work, and focused on getting basic services like email, networks, storage, and electronic result reporting rolled out to everyone in the community. Managing this work required resources, vision, and communication. All the pieces were in place to effect change.

Occasionally, I try to execute a change management project more quickly than usual, bypassing these steps. Whenever I do that I find that adoption of the new technology is delayed, budgets are at risk of overrun, and frustration escalates.

My decade of experience executing change suggests that Kotter was right. Building a guiding coalition, broadly communicating the vision, and celebrating a series of short term successes really works. I've watched projects without vision, resources or communication cause pain and anxiety throughout the organization. The good news is that we now know how to execute change and it is the role of senior management to enforce Kotter's principles in every change project.

Time, Scope, and Resources

Over the past few weeks, the number of new infrastructure project requests has peaked to unprecedented levels. The usual triage mechanisms described in my previous blog entry work well for applications, but infrastructure is different. Adding a new network port, a new telephone, or a new desktop is viewed a service business that can be ordered on demand, making it very challenging to say 'no'.

The sudden surge in requests re-emphasized to me the basic law of all IT projects - timeline, scope and resources are inter-related. If scope increases, timeline or resources must increase. If timeline is shortened, scope must be decreased or resources increased. Increasing scope, shortening timeline while leaving resources constant is not possible.

Of course, we can all work harder. There are 168 hours in a week, vacations can be postponed and nights/weekends filled. This works in the short term, but is not sustainable. "Lean" and "mean" organizations pushed too hard become "bony" and "angry" organizations.

New FTEs are not typically the short term answer. Getting new positions raises expectations of delivery capacity but hiring and training new staff take resources from existing capacity, so paradoxically getting new positions actually reduces capacity for a few months.

This means there is only one short term answer for unplanned, unbudgeted, unscheduled infrastructure requests - the scope of these requests needs to be reduced/phased or the time to do them increased.

For my requests this week, I've done the following
a. Assigned my staff to develop a standard worksheet which outlines the major time limiting steps (i.e. network connections take 90 days to provision) and thus specifies the minimum lead time for building IS support for a new location
b. Negotiated a change in scope with phasing - the initial request for a high bandwidth connection and new telephone system was morphed into a low bandwidth connection and use of the existing telephone system for now.
c. Reordered priorities - previous request were placed on hold in order to service the new 'once in a lifetime' opportunities
d. Asked for new staff - with the caveat that they will not add to capacity/throughput for 6 months
e. Requested governance changes - to ensure a central committee triages and communicates infrastructure requests for new offsite locations

There is one other strategy that I could employ if this surge in requests becomes chronic. In the past, I've staffed to average work load, not peak workload. This means that staff can put in extra hours for short term urgent increases in demand. However, I may have to staff to peak so that excess capacity is always available for the continuous infrastructure tyranny of the urgent I'm a frugal guy, doing a great deal on a limited budget , so I've never built in excess capacity.

As I work through these issues, I'll keep you updated on my progress.

Grand Rounds This Week

For great links to great medical blogs check out this weeks Grand Rounds, hosted by Dr. Geek. Thanks for including EverythingHealth, Dr. Geek. The reading choices are eclectic!

New Changes in Dental Prophylaxis

The American Heart Association has published the new recommendations that will spare thousands of patients from loading up on antibiotics before a trip to the dentist. For decades certain patients have been taking antibiotics for protection before dental procedures but now the rules have changed.Infective endocarditis is a rare but life threatening condition that was thought to be prevented by

It's time to take back the internet

Today, I've received over 250 unsolicited email advertisements for IT related products, services and seminars. This is not the usual egregious SPAM (Viagra ads, Nigerian businessmen, or phoney eBay phishing emails) which are already filtered out (see my previous post). These are legitimate emails that I do not want to receive from real companies. At the bottom of these emails are the words "This is not SPAM because you have opted in to receive these emails from us or our partners". I have never opted in to receive any such emails from anyone.

My email address appears in the masthead of many of the columns I write for print and web-based media. This has led to my inclusion in just about every IT mailing list on the internet.

I've tried opting out. Removing myself from 250 mailing lists each day can take an hour, so that's clearly not a sustainable part of my schedule. Also, some unsubscribe links just do not work. What are our options?

Short term, I advocate that CIOs start blacklisting the email domains that pummel us with unsolicited IT advertising. I'm a kind and gentle soul, so I'm willing to give companies a chance to redeem themselves. This month, I'm responding to these companies with a polite request to never send me emails again. If they continue to fill my inbox, I will blacklist their domain on my local email servers, preventing delivery of their advertising throughout my enterprise.

Thus far, I have received courteous responses from my first few companies i.e.

"Thank you for your response. Receiving 250 business solicitations each day is too much of a burden. I will happily remove your name. Your feedback is appreciated"

Long term, I'd recommend a national "do not call list" for email just as we use to discourage telemarketers from making unsolicited calls to households which have opted out on the national list. I would happily submit my name to a national email opt out list.

Other long term alternatives - we could create local whitelists of companies we do business with, but maintaining such a whitelist is challenging for companies and individuals. Other possibilities include expansion of Spam legislation and enforcement which provides sanctions for violators of unsolicited email rules.

I'm ok with "pull" advertising on websites that is customized to me, but is easy to ignore. However, I consider "push" unsolicited email contact an invasion of a personal communication vehicle.

Thus, for companies that send out endless unsolicited emails, watch for my responses begging you to stop sending me email advertising! The only recourse I have left is to block your domain if you continue.

Safe Driver Discounts for Technology

Automobile insurers have long seen the sense of giving drivers an incentive, in the form of safe-driver discounts, to avoid taking risks when they’re behind the wheel. In health care, more and more payers are rewarding doctors for the quality of care they deliver and not the quantity. Aligning incentives with outcomes makes good sense.

Vendors of technology should follow those examples and revise their pricing models for yearly hardware and software maintenance contracts. If they rewarded customers who adhere to best practices, they would essentially pay customers for their performance.

As a CIO of multiple companies, I have to sign off on a lot of maintenance contracts every year. These contracts have a list price, and there’s usually a discount that the manufacturer passes along to the value-added reseller (VAR). The VAR decides how much of the discount to pass along to customers. No extra consideration is given to customers who abide by the vendor's best practices for the implementation and management of its products.

But why not? Why not give technology buyers the equivalent of a safe-driver discount? If customers were given incentives to hire highly competent internal staff, follow all the vendor-recommended configurations and install all the latest upgrades, life would be better for both the vendor and the customer. The vendor would receive fewer support calls and requests for emergency priority service. The customer would get higher reliability, better performance and lower maintenance costs.

Five years ago this month, we experienced a devastating network outage that led me to change a lot of our practices. Before the outage, my only incentive to adopt best practices was fear of downtime. The cost of support was certainly not a factor. We could make hundreds of support calls and send out an SOS during numerous high-priority emergencies, and the cost would be the same as it would be with a spotless performance record. The outage led me to replace much of our infrastructure, enhance my support team and ensure that our engineering practices are world class. We now place an extremely low burden on our vendors, but our maintenance discounts for all the technology we operate today don’t reflect that.

Here’s my idea. Vendors would give each customer a yearly technology safety rating, starting at, say, 100 points. If you miss an upgrade, 10 points would be deducted. Deviate significantly from a recommended configuration and you lose another 10 points. Make a support call that’s due to your lack of appropriate IT staffing and more points are taken away.

Discounts would no longer be arbitrary. Instead, they would be a direct function of the yearly safety rating. The harder a customer worked to avoid calls for help, the less the maintenance would cost. The best customer of the year could even be rewarded with completely free maintenance.

Of course, there are potential problems. Vendors could abuse the system by defining best practices as the elimination of all competing products, or they could take away points if customers didn’t buy all the optional add-on software they recommend. But such tactics would defeat the spirit of this proposal. Any vendor that adopted them could expect a good deal of pushback from customers; hopefully, they would see that such transparently cynical ploys have no real value.

So let me publicly ask my good friends at Cisco, EMC, Dell, HP and IBM, What do you think? You'll find that my driving record is exemplary.

RBRVS-For Primary Care No Good Work Goes Unpunished

The results are in - Medical Residents are not choosing to enter the Generalist fields of medicine in the United States. With current trends, less than 10% of those training in Internal Medicine will work as general internists. Compare this with JAMAs report that all European countries have a broad Generalist foundation comprising 70-80% of practicing physicians. The consequences of our