News

Effective resourcing

When it comes to managing resources effectively in a CRO, the most difficult thing is finding a way to balance your portfolio of work.

On the face of it, that may sound deceptively straightforward – you pitch for work, work comes in (or doesn’t) you do the work and then you move on – but in reality, it’s much more complex than that. Over the years we’ve found that there are multiple factors that determine whether a company is able to manage resources effectively and we’d like to talk a bit about what we believe are the most important ones.

However, before we look at what needs to go right, let’s look briefly at what can sometimes go wrong in any particular project:

  • Under-resourcing – so they’re either delivered late or to an unacceptable standard
  • Over-resourcing – so that while they’re delivered on time and to your clients’ satisfaction, the business isn’t sustainable
  • Cross-resourcing – such that both the client and the CRO end up duplicating (and wasting) effort and expertise by carrying out the same tasks
  • Unfocused-resourcing – lack of clarity on roles and responsibilities leading to essential tasks not being completed by either party

So, how best to ensure that the project pipeline delivers the right results by striking the proper balance between bidding work and contracted work and then reviewing that balance on a regular basis?

We believe there are certain key drivers involved in getting this right.

How enduring partnerships grow from good governance

There are multiple factors that determine whether a company is able to manage resources effectively

Visibility

Like driving safely through fog, the more visibility you have, the better you’ll be able to negotiate the twists and turns of the road ahead, and also avoid any unanticipated obstacles. This starts by having business development and operational staff working hand-in-glove to quote for new business, so there’s confidence that teams won’t be over-stretched and that projects will be delivered on time.

Project managers are best placed to be in charge here, because the environment is so dynamic (even effective teams can rarely predict what they’ll be working on more than one month ahead) that they need to be hands-on, so they can react as projects get postponed or cancelled, and re-allocate resources elsewhere. They also have an important role to play in sanity-checking projects and need that finger-on-the-pulse familiarity with each piece of work so that if a regulatory matter crops up, they can respond at speed and with the necessary level of detail.


Agility

It’s said that in Amazon’s early days, founder Jeff Bezos based his management structure on small teams – no more people than could be fed on two pizzas. Takeaway preferences aside, it’s important that people are assembled into appropriately-sized teams that can re-organise themselves as necessary during the life of a project. While you don’t want people jumping on and off teams for the sake of it (and the most successful projects tend to be those that have the same team from concept to completion) you do need to be agile enough to understand that a project may need to be resourced flexibly – for example, one person at the start, 10 in the middle, but only two at the end. Similarly, if a team delivers a project early – you need to be able to identify opportunities for them to be reallocated quickly as required.

It’s important that people are assembled into appropriately-sized teams that can re-organise themselves as necessary during the life of a project

Fluidity

You have to accept – and where possible, anticipate – that resourcing isn’t static. It’s fluid. It moves. Projects move around all the time, changing timelines or emphasis or importance relative to other projects, and your resources need to move with them. It might be important for example, to over-resource a specific project for the first milestone, or if it’s a tricky deliverable, or the client seems overly concerned about that particular stage, in order to achieve a better result further down the line. This kind of fluidity is fine, so long as there’s a structure in place for implementing it. What you want to avoid is the scenario where projects become like a kids’ game of football, where everyone’s running after the same ball in a massive huddle. That’s a recipe for getting nothing done.

A final thought. Successful resourcing is not just about future planning. You have to run the machine so it’s equipped to deliver consistent, high-quality work, to time – but not so lean that everybody is constantly over-worked, and not so over-resourced that delivering projects does not make business sense. Instead, you should try and plan for the most likely scenario and – using the principles we’ve outlined above – put a monitoring / management structure in place to be able to cope with alternative scenarios and fluctuations as they arise.

It may be impossible to predict the future, but that shouldn’t stop you from planning for it.

Veramed Statistical Consultancy Services

Our team of statistical consultants are experienced in providing support across the drug development lifecycle, from pre-clinical to phase IV. By contributing to the study concept we help to ensure that each study is designed to answer the scientific questions of interest by selecting the right endpoints, population and analysis methods for inclusion in the protocol and statistical analysis plan, thereby defining the appropriate estimands.

We routinely use simulations to evaluate whether novel designs, adaptive and Bayesian approaches can help to optimise the probability of study success and we work collaboratively with your clinical teams to define a framework for clear decision making throughout the study. The Veramed team aims to be flexible and innovative, focused on the needs of each individual customer to achieve the best possible study outcome, whilst maintaining the highest quality of support.

Download our consulting services infographic on the right and find more information at www.veramed.co.uk./statistical-consultancy-services/

Consultancy Services Infographic

Download

2.31 MB

Effective meetings

How many times have you emerged from a stuffy meeting room, sheaf of papers under your arm, cold cup of coffee in hand and thought to yourself: “That was a really good meeting.”

Once, twice, most of the time? You’re one of the lucky ones.

Meetings often start with the best of intentions, only to meander along, blown this way and that by pre-conceptions, poor preparation, ennui, the odd overbearing personality, until, at the appointed finish time they whimper and close; perhaps the only note of their passing, an email to ‘pick this up again next time’.

How enduring partnerships grow from good governance


The framework

But what if there was a framework to ensure that every meeting started on time, stayed on track and went on to deliver what it set out to achieve? That would be something worth remembering, and that’s why people who are ‘good’ at meetings have turned to mnemonics for help. A mnemonic is a device that aids or improves memory and usually takes the form of a memorable word or phrase that ‘spells out’ a message. Other professions – from physicians and lawyers to musicians and driving instructors – use mnemonics successfully, so why shouldn’t we?

Attendees should feel that they can contribute to the meeting and that they have defined roles and responsibilities

Our mnemonic

With that in mind, we at Veramed set out to see if we could construct a simple mnemonic that would remind us what good meetings were all about – making our company more effective, strengthening our teams, and making team engagement an enjoyable and efficient use of time. The mnemonic we arrived at was designed to be memorable (of course) and to re-enforce the belief that effective meetings are all about people working together in harmony towards a common goal.

The mnemonic acts as a reminder of the successful elements that comprise a ‘good’ meeting. Not all meetings have the same purpose, but the effectiveness of any meeting boils down to these key concepts.

A is for AGENDA. This ensures that everyone going to the meeting knows what’s going to be discussed, whether or not they need to prepare and who’s leading the meeting. It also defines the purpose of the meeting.

N is for NECESSITY. Do we need to hold the meeting now, will it start and finish on time, are the right people invited and are we able to cover all the items on the agenda?
I is for INFORM. It’s important that the right people attend – and not the wrong people. You need the personnel present who can take action on the conclusions that arise from the meeting.
C is for CHAIR. This is the person who keeps the meeting on track and who decides when to take a discussion offline, or even ‘park’ the meeting until later. They also make sure that proper minutes are taken, because without minutes, people don’t take action.
E is for ETIQUETTE. The way a meeting is conducted is vital. It starts by choosing the right chair, then giving people the opportunity to speak, respecting diverging opinions (whilst not allowing a meeting to turn into an argument or tolerating rudeness) right through to simple practicalities: for example, at Veramed we try to engender the philosophy of focusing on the task at hand – laptops only for the purpose of the meeting, mobile phones set to silent (or better, turned off).

T is for TIMELY. Has the meeting been set at the appropriate point in the project so that it’s still possible to take meaningful action that will impact the process in a positive way?
E is for ENGAGEMENT. Attendees should feel that they can contribute to the meeting and that they have defined roles and responsibilities.
A is for ACTIONS. Remember the point of the meeting is to achieve a set of clearly-stated goals.
M is for MINUTES. Accurate minutes are the unsung heroes of any effective meeting because they help turn intentions into actions. They don’t need to be long and detailed, and can just be a list of actions with person assigned and expected timeline for delivery.

The result? A NICE TEAM is a mnemonic which reminds all of us at Veramed of the components and actions that go together to make an effective meeting. Handy then, that we also think it aptly links to our Veramed values.


Feel free to adopt it in your own company if you feel it could do the same for you.

Veramed is a CRO that provides a range of high quality biostatistics and programming services for the design, reporting and analysis of clinical trials.

Please get in touch for more information.

Accurate minutes are the unsung heroes of any effective meeting because they help turn intentions into actions

What Drives Quality? – Part III

How enduring partnerships grow from good governance

We’ve argued elsewhere that delivering excellence relies on striking a balance between the ‘three pillars of quality‘ – your people, your processes and the way you support both with technology.

This is a difficult enough balance to strike when you’re talking about one or two teams delivering one or two projects for one or two stakeholders, but what happens when you upscale?

How can a CRO continue to deliver quality as the business grows and they take on more clients and projects?

How enduring partnerships grow from good governance

In order to deliver quality at this larger organisational level, we believe there are several elements that need to be put in place, specifically:

  • switching from tactical activity to strategic planning
  • moving from transactional arrangements towards a true partnership
  • demonstrating the ability to leverage information to maintain quality during periods of growth

The key to implementing these elements successfully is what’s known in legal circles as a ‘mastery of the brief’ – or in our case, a forensic understanding of the systems and principles required to oversee multiple clinical trials (sometimes with different clients) through to their successful conclusion.

We call it ‘good governance’ and we believe it’s central to preserving the integrity of the three pillars of quality

Good governance isn’t just about controlling budgets, metrics and other Key Performance Indicators (KPIs) – though these are important.

It’s also about being able to work successfully with key decision makers and influencers across teams, departments and entire organisations to build trust, and to create living, breathing partnerships that run far deeper than the traditional, often transient, business relationship.

It requires that the CRO has a granular understanding of the staffing and resourcing required to complete multiple concurrent studies, as well as the ability – and the willingness – to share those insights with clients openly and with complete honesty. This depth of collaboration furnishes the CRO and the client with the capacity required to plan for unknowns, and ensures the partnership has the necessary organisational agility to switch focus when required, to move resources in order to relieve unforeseen pinch points – even to reconfigure entire teams in order to meet the challenge of new priorities. Working this closely together allows you to manoeuvre more effectively should workloads increase unexpectedly – and to cope with the resulting rise in ‘temperature’.

Although we’ve touched on this before it’s also important to understand that good governance is about much more than following standard operating procedures (SOPs). These need to be understood and applied of course, but they must also be subject to continuing critical review so that they don’t take over and turn the process into an exercise in ticking boxes. That’s why the central plank of good governance is the ability to exercise control over projects, resources and quality across the whole range of clinical studies, whether these are current, planned or scheduled for sometime in the future.

Good governance can be the platform that gives the modern CRO the confidence to say ‘no’ from time to time – passing up on a project if the right resources are not available to deliver a quality product (rather than yielding to the persuasions of the sales team). Too few people chasing too many deliverables will create the perpetual crisis cycle; and while there’s nothing wrong with small doses of timely pressure, constant stress will inevitably result in a loss of quality.

Good governance – encompassing oversight, escalation paths, monitored feedback and effective risk management – applied across the project and client portfolios and at every level allows today’s CROs to embrace new and expanding collaborations with confidence. That’s because deep partnerships of the kind we’ve been describing – courtesy of good governance – enable CROs to understand what drives their clients, helps them to work in a transparent fashion that promotes trust and leads to the ultimate goal – shared values. Achieve that and the quality of your work will be exceptional, no matter how big your client base becomes.

What Drives Quality? – Part II

As we explore the central question of what drives quality one thing becomes clear – it’s easier to define what quality is than to define how you go about delivering it. That’s because ensuring excellence across a portfolio of projects demands that you’re able to harness what we call the ‘three pillars of quality’ – people, process and technology. These three elements need to come together in near-perfect harmony if you’re to deliver consistent quality across the board.

Let’s take a look at each of these in turn to see if we can define why each pillar is so important and how they can work together to deliver the right results.


People

The first pillar. Over the last 10 years or so, the industry has fallen in love with outsourcing as a way of reducing costs. This has worked to an extent, but it’s also introduced a problem, because the moment you reduce people to a commodity – a line in a budget sheet – you lose sight of all the subtler ways they can contribute to a company’s success. Things like their experience, their understanding of nuance, and their ability to work as part of a multi-functional, multi-disciplinary team.

How do you find these people? It starts from the first interview where you’re looking not just at the person in front of you, but also who they could become – tapping into their potential.

Why? Delivering quality in a small team is easy, but in a larger team or across multiple teams you need more than just highly qualified people, you need a skilled, well-rounded workforce with a questioning outlook – the data scientist mindset if you will – that approaches high volumes of data and complex derivations with a sharp eye and appraising attitude.

The People Challenge – One of your team discovers a problem. Do they try and cover it up or stare the mistake straight in the face and ask what they can do differently in order to avoid it next time around?

Process

The second pillar. Everything in our industry has a rigorous process – a set of repeatable steps tying a project together – that team members need to understand and follow. However, if the process becomes too prescriptive then there’s a very real danger that quality control becomes an end in itself and you lose sight of why the process is there in the first place.

You need good processes of course, because the industry demands it and even the best, most highly qualified people can be disorganised without a proper structure. Processes also evolve: ICH E6 Good Clinical Practice (Revision 2) came into force in late 2016 and whilst the focus is on risk-based monitoring and client oversight, it can equally be applied to support the QC and oversight of statistical reporting deliverables. Namely, checks that focus in on the most important aspects of a study – like the primary endpoint or key safety outcomes. But like all good processes it’s important to remember that these changes facilitate quality – they don’t equal quality.

The Process Challenge – A team follows a 20 step process methodically, one step after the other. Is this an exercise in box-ticking or a framework for delivering excellence?

Technology

The final pillar. In daily life, nearly everything that used to be done on paper is now done by computer, and the temptation in our industry is to do the same: employing a bit of technology to solve each and every problem. The key driver here is to understand which parts of the process should be computerised and which parts are still best performed by people. Whilst computers are good for replacing laborious tasks, overall efficiency should always be front of mind. Automating the right processes allows people to use their insight and experience to assess the data from raw data through to final deliverables. In this way, technology can be used to give your team time to do the things that machines cannot – consider, reflect, question and troubleshoot.

The Technology Challenge – Is comparing two outputs side-by-side like an old-fashioned proof reader really a good use of your team’s time?

This then, is the secret to delivering consistent quality across multiple projects. You must strike the right balance in each of the areas so that you get experienced, highly qualified people who can follow your processes intelligently – rather than slavishly – using technology in ways that frees them to stay creative and innovative, all the while applying a data scientist’s questioning mindset to their work. It’s not easy, but in terms of consistent quality, the results are well worth the effort.

Making these three pillars work together in harmony requires something that we’ve referred implicitly to throughout this article – balance.

What Drives Quality? – Part I

Quality should be the bedrock

It should be the characteristic that runs through a company from top to bottom, like lettering through a stick of seaside rock.

It should be the governing principle that defines the way a business delivers services to its customers.

Everyone knows this.

Quality should be the bedrock

So why is it so difficult for companies to deliver a top quality product or service on a consistent basis?

General Principles

When you’re considering the principles that drive quality, it’s useful first to think in general terms. So you should:

  • establish clear standards that everyone understands
  • find the right people with the right mindset – and qualifications – to implement those standards
  • be vigilant to ensure those standards are adhered to
  • lead your team in avoiding ‘quality fatigue’
  • re-connect with core quality values periodically to make sure there’s been no drift

The commitment to delivering quality must be broad and deep, it must involve both your people and your processes and be the key motivator for everyone in the company

– from those who look after the big picture to those who pore over the individual data points that make the business tick.

Let’s look at how we might apply these general principles to the biostatistics and programming aspects of the pharmaceutical business.

The Keys to Quality

Quality needs to be quantifiable so that you can assess whether you’re meeting your own benchmarks, but it can mean different things to different people. In a pharmaceutical trial for example, the various stakeholders are likely to have contrasting views as to where the emphasis should lie, depending on whether they’re the CRO, the client or the auditor.

From the CRO point of view, quality means having the processes and the people in place that allow the organisation to get the results right, first time, every time; that’s the foundation upon which everything else depends.

The client on the other hand, demands all that plus first class communications, transparency (particularly in terms of planning and risk management) and reassurance in the form of regular meetings or updates.

Then there’s a third, competing demand because if the auditor comes calling then detailed documentation needs to be in place to satisfy a rigorous examination and to demonstrate proper process and adherence to Good Clinical Practice (GCP).

Quality In-House

Building a team that can deliver consistently high quality results is a bit like putting together a complex jigsaw puzzle, because the people involved must have the right combination of skills, qualifications and experience.

At the same time, the team must understand exactly what’s expected in terms of quality and having established the ground rules, management needs to re-visit this periodically.

Forensic attention to detail of all available documentation such as the study protocol, the Statistical Analysis Plan and the dataset specification – even to the Table Shells themselves – is vital to ensure consistently high standards are maintained.

When you’re able to work at that level of detail, everyone in the team understands exactly what’s required – right down to the necessary number of decimal places in a specific statistic.

Quality In Relationships

Although it may seem counter-intuitive, it’s possible to produce perfect deliverables, on time, and still end up with a client that isn’t happy.

That’s because it’s the quality of the relationship that keeps clients happy and drives repeat business.

If a client has doubts or feels uncertain that their CRO of choice will be able to deliver a project on time or to the required standard, they’re going to lose trust even if you hit every milestone with quality results.

This is where transparency comes in. Working together in a collaborative manner promotes respect within the client relationship along with an important sense of shared responsibility.

Quality For Auditing

“If something isn’t documented, it didn’t happen.”

If your processes are in place, your staff have the right skills matrix and your governance is strong, then an audit will merely confirm that your study conduct, performance, monitoring, analysis and reporting meet the required international standards of GCP.

Understanding how this fusion of world class technical documentation and a good client relationship go hand in hand in driving overall quality is crucial in terms of delivering a great service to your clients, and doing it consistently, time after time.

Veramed is a CRO that provides a range of high quality biostatistics and programming services for the design, reporting and analysis of clinical trials.

Please get in touch for more information.

“Rescuing” studies

“Rescuing” Clinical Trials: From poor delivery to high performance

No client goes into a clinical trial with the expectation that their CRO of choice is going to let them down at a critical point, and that they may have to re-examine elements of the study or – in the worst case scenario – start again.

But with so much penetration into the clinical process by CROs across the board (70 per cent of studies by 2020 according to some) it’s almost certain to happen at some point. Learn about the warning signs, and how to move forward with your study.

Rescue Study

Download

477.07 KB

What does it take to become an Associate Statistician?

In our latest resource post, Statistician, Ingrid Franklin imparts her knowledge on how she came to join Veramed and her experience working at Veramed.

After 4 years studying for a BSc and with graduation fast approaching, I, like many others, still had no idea what I wanted to do when I entered the ‘real world’ of employment. I jumped at the first job offered to me, worked as an Economist for a year and quickly realised that I was unhappy and that this was not the job for me.

veramed_e_p

Completing a Masters

Thinking back to what I had enjoyed as an undergraduate, a medical statistics module sprung to mind. I knew I needed to do something about my situation and so headed back to university to streamline my studies, embarking on an MSc in Statistics with Medical Applications, in the hope that this would lead me into a career that I was better suited to.

The course was not easy, but I found the fact it had a true focus made it more interesting to me than much of what I had learnt during my undergraduate degree. The one year course flew by, and before I knew it, it was graduation time again! Not wanting to make the same mistake twice, I didn’t rush to look for a job at the earliest opportunity; instead, I opted to start job hunting towards the end of the academic year, once I had had more time to research what was available.

Doing lots of research into the industry

The PSI careers fair (usually in February) was a fantastic opportunity to chat to a range of industry statisticians, from those that were just a year ahead of me, right through to people with 20+ years of experience. There were introductory talks on becoming a statistician and exercises for all attendees, as well as the opportunity to wander around the stands at your leisure to try and suss out the options available to a young medical statistician.

Enthused by the people I had been speaking with, I was now excited to get back into the world of employment. After all I had heard and researched on becoming a statistician, I decided I wanted to work for a CRO as it seemed that this would provide me with a good variety of projects to work on. I was also naturally swayed towards a smaller organisation, as I knew this would give me the chance to be a very valuable member of the team and would hopefully mean I was exposed to a vaster range of subject areas to work in.

My experience working for Veramed

I have now worked for Veramed for over two years, and was absolutely right in my assumptions. I began working as an associate statistician on early phase trials that were mainly immunology based. My everyday tasks included reviewing many study documents, from protocols to study reports and statistical outputs, as well as writing statistical analysis plans and attending study meetings. This exposure to the overall study process meant my knowledge of the everyday workings of a clinical trial rapidly expanded. I also learnt to program in SAS, largely using it to produce descriptive data outputs or to run analysis on trial data.

One of my favourite aspects of working on early phase trials is being involved in the initial research and design. The trials I work on are often for rare diseases and I always find learning about these fascinating, as I gain a real insight into the underlying foundations for the study. There is often a lack of historical literature available for these rare disease areas and so I do feel my research skills and ability to find useable and relevant data needed for a design have greatly improved. Study designs I have worked on range from those that use a basic frequentist methodology to more technical Bayesian designs, which are becoming more and more common in the industry. After building up a wealth of experience so far, I am now in the position of being able to be the lead statistician for trials, as well as work on higher level projects that relate to the development of a particular compound for a number of studies, rather than simply working on studies as standalone assignments.

Apart from working on designs and ongoing clinical trials I have been able to work on veterinary studies too, which has been a learning curve that I think will be beneficial because the process and regulations are different to that of a clinical trial. I’ve also recently been lucky enough to work on biomarker projects, which is an exploding area of interest in the pharmaceutical industry, where the modelling of genetics data is used to predict whether a patient is likely to respond to a particular treatment.

Attending industry conferences

In my time at Veramed I have attended the PSI conference twice; this provides an excellent chance to meet other statisticians in the industry and to learn about new statistical techniques and current hot topics. Presentation objectives range from increasing awareness of the latest regulatory guidelines, to exhibiting the new and ever adapting techniques used to improve processes and combat common problems in clinical trials. I have also been able to attend multi-day statistics training courses to ensure I have an up-to-date and in-depth knowledge of methods used in my everyday work.

In summary, the two years I have spent in industry so far have been both interesting and challenging. My team is encouraged to share knowledge, which is extremely useful in a CRO because people will be constantly gaining unique experiences depending on their particular projects and client relations. In my current role I feel that I have been continuously able to develop, through working on a wide variety of projects, while taking on more responsibility over time. I know that challenges in the coming year will ensure that I am able develop professionally further still and I am very much look forward to doing so.

Find out about available positions with Veramed

Do your SAS comments add value?

The below is a short abstract from Veramed’s Good Programming Practice guidelines. Download the full guide using the link on the right.

The  purpose of this Guideline (GUI) is to provide guidance for good programming practices (GPP) for analysis, reporting and data manipulation of clinical data in health and life sciences organisations.  This guidance is primarily aimed at programming in SAS, however the principles of GPP also apply to other languages such as R and Stata.  In addition, although this is not produced with SAS macros in mind, the same principles apply to macros too.

We often have to update existing programs to add new rules, copy programs from one study to another, and take over programs written by others.  The guidance aims to show how to produce well structured and well documented programs so that they are easy to read and maintain over time.  It is meant to be applicable to all programs, and hence all programmers regardless of experience.  Specific rules may be of more use to novice programmers, but applying the principles should be in mind for experienced programmers and mentors.

Comments

Comments are important to help anyone reviewing, modifying or using a program to be able to quickly understand the code. All major data or proc steps should be commented, especially data specific and complex code. Ideally comments should be comprehensive, and should describe the rationale and not simply the action. For example, instead of simply typing “Access demography data”, describe which data elements you are accessing and why they are needed, for example, “Bringing in DM to get gender and age and subset to include only the intent to treat population”. Comments can also include links to external documentation (requirement specifications, design documents. The programs should also be split up into sections by creating different types of comments, e.g. many rows with asterisks. This helps to structure the program and make it easier for others to see an overview of the program.

Further guidance from Veramed
  • Consider the most appropriate method of commenting for the placement of the comment. For example macro comments should be used in macro code, block comments should be avoided in favour of line comments.
  • Use a consistent style of commenting throughout the program.
  • Write comments as if someone new to the project was looking at the code.
  • Avoid punctuation marks that may cause syntax errors. Grammatically incorrect English is acceptable!
  • Include comments at appropriate places in the program in order to aid clarity.
  • Explain subsetting.
  • Identify the activity being performed.
  • Explain the reasons behind the coding.
  • Ensure comments are consistent with the code they are explaining. If the code is updated, ensure the comments are too.

Good Programming Practice Guide

Good Programming Practice Guide

2.4Mb

Do you practice safe SAS?

Do you practice safe SAS?
The below is a short abstract from Veramed’s Good Programming Practice guidelines. For the full guide please download using the link on the right.

The  purpose of this Guideline (GUI) is to provide guidance for good programming practices (GPP) for analysis, reporting and data manipulation of clinical data in health and life sciences organisations.  This guidance is primarily aimed at programming in SAS, however the principles of GPP also apply to other languages such as R and Stata.  In addition, although this is not produced with SAS macros in mind, the same principles apply to macros too.

We often have to update existing programs to add new rules, copy programs from one study to another, and take over programs written by others.  The guidance aims to show how to produce well structured and well documented programs so that they are easy to read and maintain over time.  It is meant to be applicable to all programs, and hence all programmers regardless of experience.  Specific rules may be of more use to novice programmers, but applying the principles should be in mind for experienced programmers and mentors.

Log File Checking

As part of development and validation practices, it is often mandated that the log file generated is checked to ensure that the program has executed in the correct intention.  Many companies may have their own automatic log file checking utilities to aid in this, and there are many examples of such tools in widely available papers.  “ERROR” and “WARNING” in logs should normally be avoided.  There are sometime exceptions to this, such as warnings that are output from statistical models that do not have enough data.  Ordinarily, any warnings that are deemed acceptable are documented.  There are also some specific “NOTE”s that can indicate a problem.  The common “NOTE”s that should normally be avoided include those relating to “repeats”, “more than one”, “uninitialized” and “referenced”.

Also, any user defined checks that have been added, such as from defensive programming, should be checked for in the log and followed up on.  A company-specific naming convention for user defined checks can aid in this, so the specific string can be searched for within the log.  Examples of such conventions include “ISSUE:”, “USER:”, and “ALERT:”. Avoid the use of user-generated errors and warnings labelled “NOTE:”, “WARNING:” or “ERROR:”, as these may make it difficult to find genuine problems when searching the log.

Good Programming Practice Guide

Good Programming Practice Guide

2.4Mb